Proteomics Reveals Ablation of Placental Growth Factor Inhibits the Insulin Resistance

1 Proteomics reveals ablation of placental growth factor inhibits the insulin resistance

2 pathways in diabetic mouse retina

3 Madhu Sudhana Saddala1#, Anton Lennikov1# , Shibo Tang2,3, Hu Huang1,2,3* 4

5 1 Wilmer Eye Institute, Johns Hopkins University, Baltimore, Maryland, United States of America

6 2 Aier School of Ophthalmology, Central South University, Changsha, Hunan, China

7 3 Aier Eye Institute, Changsha, Hunan, China

9 *Corresponding author:

10 Hu Huang, PhD

11 Wilmer Eye Institute, Ophthalmology-Retinal Vascular Service

12 400 North Broadway St. Baltimore, MD, 21287

13 Hospital M017 Smith Building

14 Tel +1 (410)5020807

15 [email protected]

17 # Madhu Sudhana Saddala and Anton Lennikov have contributed equally to this work.

19 ORCID

20 Author 1: Madhu Sudhana Saddala ORCID: 0000-0002-6373-7080

21 Author 2: Anton Lennikov ORCID: 0000-0001-8625-1211

22 Author 3: Shibo Tang ORCID: 0000-0003-2737-6780

23 Author 4: Hu Huang ORCID: 0000-0003-2843-0320 24

1 bioRxiv preprint doi: https://doi.org/10.1101/338368; this version posted June 4, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

25 Acknowledgments

26 The authors wish to acknowledge the contribution of: Lijuan Fan for technical assistance, Dmitry Rumyancev for

27 artwork design, and Jianjiang Hao for MS analysis.

28 Funding:

29 This work was supported by NIH grant (EY027824)

30 Authors Contributions:

31 The study was conceived and designed by M.S.S., A.L. and H.H. H.H. performed the animal handling, sample

32 collection and in vivo examinations. The manuscript was written by M.S.S, A.L.,H.H. and critically revised by H.H.

33 and S.T. All Authors reviewed and accepted the final version of the manuscript.

34 Additional Information

35 Financial interest statement:

36 The authors have no financial interests to disclose in relation to this paper.

38 Conflict of interest statement:

39 The authors have no conflict of interests to disclose in relation to this paper.

2 bioRxiv preprint doi: https://doi.org/10.1101/338368; this version posted June 4, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

41 Abstract

42 The underlying molecular mechanisms that placental growth factor (PlGF) mediates the early complications at non-

43 proliferative diabetic retinopathy (DR) remain largely elusive. The objective of this study is to characterize

44 expression profile due to PlGF ablation in the retina of diabetic mice. The quantitative label-free proteomics was

45 carried out on retinal tissues collected from mouse strains (Akita; PlGF-/- and Akita.PlGF-/-). We have identified

46 3176 total proteins, and 107 were significantly different between the experimental groups, followed by gene

47 ontology, functional pathways, and protein-protein network interaction analysis. Gnb1, Gnb2, Gnb4, Gnai2, Gnao1,

48 Snap25, Stxbp1, Vamp2 and Gngt1 proteins are involved in insulin resistance pathways, which are down-regulated

49 in PlGF ablation in Akita diabetics (Akita.PlGF-/- vs. Akita), up-regulation in Akita vs. C57, PlGF-/- vs. C57. Prdx6,

50 Prdx5 (up-regulation) are known of antioxidant activity; Map2 is involved in neural protection pathways which are

51 up-regulated in Akita.PlGF-/- vs. Akita. Our results suggest that inhibition of insulin resistance pathway and the

52 enhancement of antioxidant defence and neural function may represent the potential mechanisms of anti-PlGF

53 compounds in the treatment of DR.

54 Total words: 175

59 Keywords: Diabetes, Akita mice, PlGF-knockout, Mass spectrometry, proteomics

3 bioRxiv preprint doi: https://doi.org/10.1101/338368; this version posted June 4, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

61 1.0 Introduction

62 Diabetic retinopathy (DR), a sight-threatening microvascular complication of diabetes myelitis (DM), remains the

63 leading cause of vision loss worldwide in the adult population, especially in economically developed countries. (Lee

64 et al, 2015) With the increasing number of people with DM, the prevalence of DR and diabetic macular edema

65 (DME) is expected to grow. (Wild et al, 2004) Metabolic changes in the diabetic retina result in the altered

66 expression pattern of some mediators including growth factors, neurotrophic factors, cytokines/chemokines,

67 vasoactive agents, and inflammatory and adhesion molecules, resulting in vascular lesions and cell death. (Chen et

68 al, 2015; Kowluru & Mishra, 2015; Liu et al, 2015) Emerging evidence suggests that retinal neurodegeneration is an

69 early event in the pathogenesis of DR which could participate in the development of microvascular abnormalities.

70 (Barber, 2015; Simo et al, 2014) Placenta growth factor (PlGF), a member of VEGF family proteins, first discovered

71 in human placental cDNA in 1991.

72 In over two decades of scientific research and development have increased our understanding of the PlGF biological

73 function. Despite the high level of expression in placenta, the ablation of PlGF in mice did not compromise the

74 healthy embryonic development or adverse postnatal health effects. (Carmeliet & Jain, 2011) Delivery of

75 recombinant PlGF homodimer, PlGF-VEGFA heterodimer significantly promoted angiogenesis in ischemic

76 conditions through FLT1. (Luttun et al, 2002) Furthermore, many other cell types express PlGF in pathological

77 conditions, including retinal pigment epithelial cells (RPE). (Hollborn et al, 2006) This upregulation is due not

78 only to hypoxia but also from stimulus including nitric oxide (Mohammed et al, 2007), cytokines, as interleukin

79 1 (IL-1) and TNF-α (De Ceuninck et al, 2004), and transforming growth factor-β1 (TGF-β). (Yao et al, 2005)

80 The observation further confirmed the specific role of PlGF in pathological conditions that during pathological

81 angiogenesis endothelial cells over-express PlGF. (Ponticelli et al, 2008)

82 Recently our group has reported PlGF deletion in C57BL/6-Ins2Akita/J (Akita) mouse line, containing a dominant

83 mutation that induces spontaneous diabetes with a rapid onset. (Barber et al, 2005) Ablation of PlGF in the diabetic

84 mice resulted in an decreased expression of diabetes-activated hypoxia-inducible factor (HIF)1α, vascular

85 endothelial growth factor (VEGF) pathway, including expression of HIF1α, VEGF, VEGFR1–3, and the extent of

86 phospho (p)-VEGFR1, p-VEGFR2, and p–endothelial nitric oxide synthase, in the retinas of diabetic PlGF−/− mice.

87 Without a noticeable effect on glucose balance or expression of intercellular adhesion molecule-1, vascular cell

88 adhesion molecule-1, CD11b, and CD18 (Huang et al, 2015a).

89 While many functions and biological roles of PlGF are still currently unknown, the transition to human patients with

90 two phase II clinical trials of anti-PlGF recombinant monoclonal antibody in human DR patients is currently

91 underway. (NCT03071068; NCT03499223; ThromboGenics, 2018) Use of PlGF antibodies in humans presents a

92 challenge of better understanding the functions and pathways involved in PlGF knock out on the proteome scale.

93 Label-free mass spectrometry (LFMS) is a widely used tool for protein identification and quantification it is a gel-

94 free method allowing to conduct whole proteome analysis without the use of isotopic labeling (Luber et al, 2010).

95 Furthermore, as DR affects the expression of many commonly used “housekeeping” proteins such as ACTB and

96 Tubulin, MS approach resolve this issue as proteins are identified by the number of peptide sequences rather than

97 any relative quantification that requires a “housekeeping” protein (Rocha-Martins et al, 2012), (Li & Shen, 2013). In

98 the current study, we used label-free quantitative proteomics analysis to study retinal protein extracts from three

99 genetically modified mouse strains with diabetic and PlGF knockout condition as well as a combination of both to

100 further elucidate the molecular mechanisms of PlGF knockout playing beneficial role in DR on the proteome wide

101 level. We have identified 3176 total proteins, and 107 were significantly different between the experimental groups

102 (p<0.05).

103

104 1.0 Results

105 2.1 Animals and diabetic conditions.

106 Four 5-6 months old female mice were selected from each strain: C57BL/6-Ins2/J.PlGF-/- (Akita.PlGF-/-),

107 PlGF-/-, C57BL/6-Ins2/J (Akita), and C57BL/6J (C57) for this study. Graphical abstract of the animal

108 breeding program is presented in Figure 1A. Animals blood glucose, levels of glycated hemoglobin (HbA1c) and

109 body weight are presented in Table 1. All Akita mice have demonstrated significant increase of blood glucose (BG)

110 levels (p<0.001), HbA1c (p<0.05) and a decrease in body weight (p<0.01) when compared with C57 control animals

111 of the same age. Lack of PlGF did not affect blood glucose (p>0.05) levels of glycated hemoglobin (HbA1c)

112 (p>0.05) and body weight (p>0.05) for Akita.PlGF-/- vs. Akita. There was no significant difference in any of the

113 parameters in PlGF-/- vs. C57 mice of the same age. Retinal protein extracts 4 samples per group were trypsin

114 digested and subjected to the LC/MS/MS analysis (Figure 1C,D).

115

116 2.5 Label-free proteomics data analysis

117 The MaxQuant framework is used for proteomics data analysis, which is written in C# in the Microsoft .NET

118 environment. Algorithmic sets of MaxQuant are freely accessible as source code, and the complete program can be

119 downloaded from www.maxquant.org web link. We followed detailed instructions for installation and support

120 programs Cox et al., 2009. The four experimental data sets (Akita.PlGF-/-, PlGF-/-, Akita, and C57) were taken as a

121 raw file (Table 1) for label-free quantification using MaxQuant version 1.6.01 (http://maxquant.org/). Graphical

122 abstract in Figure 2 demonstrates the overall analysis workflow. For quantification results of the dataset, MaxQuant

123 calculated the number of quantified proteins, peptides, and sites. The total 200102 msmsScans, 88071 msScans,

124 5990710 peaks, 465 oxidation(M)Sites, 668368 peptides and 794 proteins were identified (FDR < 0.01) along with

125 mass, m/z, scans, and peaks, etc., for all data sets in the combined/text folder of the MaxQuant output directory.

126

127 2.6 Perseus analysis pipeline

128 Perseus is software for shotgun proteomics data analysis, which helps to extract biologically meaningful information

129 from processed raw files. It performs Bioinformatics analyses of the output of MaxQuant, and thus completes the

130 proteomics analysis pipeline. This allows easy integration of an unlimited number of independent statistical tools,

131 which can thus be combined in an analysis. The software already includes various statistical methods and

132 illustrations, such as data transformation, normalization and imputation, unsupervised and supervised learning

133 methods, correlation profiling, enrichment tests, motif identification, volcano plots, scatter plots and more. The

134 MaxQuant software generated output file “proteingroups.txt” was used for statistical analysis by Perseus framework

135 version 1.6.1.1 (Tyanova et al, 2016). The four experimental samples (each one has four individual replicates) were

136 taken as four combinations like Akita.PlGF-/- vs Akita, Akita vs C57, PlGF-/- vs C57 and Akita.PlGF-/- vs PlGF-/-

137 respectively for further statistical analysis. We have estimated the LFQ intensity of all combinations and intensity

138 average is at the same level to the all samples, which describe that the results of LFQ intensity analysis have no

139 biases to the all samples. The correlation coefficient of LFQ intensities was higher than 0.839 in biological

140 replicates. In Akita.PlGF-/- vs Akita combination 0.839 low LFQ intensities between Akita.PlGF-/- replicate 1 and

141 Akita replicate 3, 0.937 high LFQ intensities between Akita.PlGF-/- replicates 1 and 2. (Suppl. Figure 1B, C, D) In

142 Akita vs C57 combination 0.867 low LFQ intensities between Akita replicate 4 and C57 replicate 2, 0.943 high LFQ

143 intensities between C57 replicates 3 and 4. (Suppl. Figure 2B, C, D) In PlGF-/- vs C57 combination 0.876 low LFQ

144 intensities between PlGF-/- replicate 1 and C57 replicate 3, 0.945 high LFQ intensities between PlGF-/- replicates 2

145 and 4 (Suppl. Figure 3B, C, D). In Akita.PlGF-/- vs PlGF-/- combination 0.877 low LFQ intensities between

146 Akita.PlGF-/- replicate 1 and 3, 0.948 high LFQ intensities between PlGF-/- replicates 2 and 4 (Suppl. Figure 4B, C,

147 D), suggesting that the experiment has high repeatability and reliability. The hierarchical clustering and principal

148 component analysis showed clear separation among various components. (Suppl. Figure 1A, 2A, 3A, 4A) Based on

149 the threshold like fold-change (FC) of 2 and p-value (p < 0.05) screening differentially expressed proteins (DEPs) in

150 all combinations mentioned above. (Suppl. Tables 1, 2, 3, 4). We have also provided MS/MS spectrum and the

151 confidently identified peptide components of Gnb1, Gnb2, Prdx6 and Map2 proteins (Suppl. Figure 5A, B, C, D).

152 The Akita.PlGF-/- vs. Akita combination has 31 differentially expressed proteins (6 up-regulate, and 25

153 downregulate). Prdx6, Map2, Tubb6, Crocc, Hsp90ab1 and Ckb proteins are up-regulated and Gnb2, Snap25,

154 Pcbp1, Atp1a1, Pcbp3, Gnao1, Atp2b1, Atp5o, Gnai2, Vim, Slc25a22, Hnrnpa1, Hist1h1c, Cct8, Hist1h1e,

155 Hist1h1d, Hnrnpc, Atp6v1e1, Hnrnpul2, Mecp2, Hist1h1b, Hnrnpab, Slc6a11, Hnrnpd and Hist1h1a proteins are

156 down-regulated in Akita.PlGF-/- group when compared to Akita. (Figure 3A) The Akita vs. C57 combination has 26

157 DEPs, 15 proteins are up-regulated, and 11 proteins are down-regulated. Hnrnpd, Hnrnpc, Hnrnpab, Atp1b2,

158 Hnrnpa1, Lmnb2, Atp5o, Gnb1, Pcbp1, Gnb2, Atp6v1a, Atp1a2, Atp1a1, Hnrnpk and Stxbp1 proteins are up-

159 regulated and Glul, Cfl1, Prdx6, Hba2, Map2, Epb41l3, Ntm, Uqcrq, Gfap, Nme1 and Nefm proteins are down-

160 regulated in Akita group when compared to C57 (Figure 4A). The PlGF-/- vs. C57 combination has 31 DEPs, 15

161 proteins are up-regulated, and 16 proteins are down-regulated. Ndufa8, Cirbp, Rbp3, Napb, Crocc, Gngt1, Gnat1,

162 Atp1b2, Gnb1, Atp5o, Lmnb2, Atp5f1, Gnb4, Gnb2 and Aco2 proteins are up-regulated and Nono, Tubb3, Vamp1,

163 Ywhab, Srsf1, Hnrnpc, Hba2, Ywhag, Epb4.1l3, Tpm1, Eif4a1, Rac1, Gfap, Hnrnpa0, Nefm and Rhobtb3 are down-

164 regulated proteins in PlGF-/- group when compared to C57 (Figure 5A). The Akita.PlGF-/- vs. PlGF-/- combination

165 has 19 DEPs, 6 proteins are up-regulated, and 13 proteins are down-regulated. Tubb6, Tubb3, Hsp90ab1, Tubb2b,

166 Tubb5, and Hspa8 are up-regulated Atp5o, Map1b, Immt, Hnrnpm, Vdac2, Pura, Cct8, Cirbp, Cox5b, Mecp2, Rs1,

167 Ndufa8 and Erh proteins are down-regulated in Akita.PlGF-/- group when compared to PlGF-/- (Figure 6A). In all

168 combinations of DEPs are used for gene annotation (GO) and pathways analysis.

169

170 2.7 Functional classification and pathway analysis

171 The functional analysis is a crucial factor step in data analysis for no functional annotation of the data sets. The

172 DEPs were connecting to at least one annotation term each within the molecular function (MF), biological process

173 (BP) and cellular component (CC) classes. In this study, DAVID Bioinformatics Resources 6.8

174 (https://david.ncifcrf.gov/) and GO Enrichment Analysis (http://geneontology.org/page/go-enrichment-analysis) are

175 used for gene annotation of DEPs. All the up and down regulated individual combinations of DEPs were uploaded

176 to the DAVID annotation tool using complete mouse proteome as background. The GO terms were predicted based

177 on Expression Analysis Systematic Explorer (EASE) < 0.1 and threshold count (TC) ≥ 2. The molecular functions,

178 biological process, cellular components protein classes, and pathways were predicted in the significant enriched GO

179 terms of up and down-regulated proteins.

180 The Akita.PlGF-/- vs PlGF-/- combination DEPs were involved in different biological processes like biological

181 regulation (GO:0065007), cellular component organization or biogenesis (GO:0071840), cellular process

182 (GO:0009987), localization (GO:0051179), metabolic process (GO:0008152), multicellular organismal process

183 (GO:0032501) and response to stimulus (GO:0050896) respectively (Figure 2B). Gnao1, Gnai2, Atp2b1, Atp1a1,

184 Snap25 and Slc6a11 are involved in biological regulation (GO:0065007), Hist1h1d, Tubb6, Cct8, Map2, Hist1h1c,

185 Hist1h1a, Snap25, Hist1h1b and Hist1h1e are involved in cellular component organization or biogenesis

186 (GO:0071840), Slc25a22, Gnao1, Atp2b1, Pcbp1, Hist1h1d, Gnai2, Tubb6, Cct8, Map2, Prdx6, Gnb2, Hist1h1c,

187 Hist1h1a, Pcbp3, Snap25, Hist1h1b, Atp1a1, Slc6a11, Crocc and Hist1h1e are involved in cellular process

188 (GO:0009987), Pcbp1, Map2, Pcbp3 are involved in localization (GO:0051179), Slc25a22, Gnao1, Hsp90ab1,

189 Atp2b1, Pcbp1, Hist1h1d, Gnai2, Cct8, Ckb, Prdx6, Hist1h1c, Hist1h1a, Pcbp3, Hist1h1b, Atp1a1 and Hist1h1e are

190 involved in metabolic process (GO:0008152), Pcbp1, Map2, Gnb2, Pcbp3, Snap25 and Slc6a11 are involved in

191 multicellular organismal process (GO:0032501), Gnao1, Hsp90ab1, Gnai2, Mecp2, and Prdx6 are involved in

192 response to stimulus (GO:0050896) respectively. The DEPs were involved in different molecular functions like

193 antioxidant activity (GO:0016209), binding (GO:0005488), catalytic activity (GO:0003824), signal transducer

194 activity (GO:0004871), structural molecule activity (GO:0005198) and transporter activity (GO:0005215) (Figure

195 2C). Prdx6, Prdx5 have antioxidant activity, Gnao1, Atp2b1, Pcbp1, Hist1h1d, Gnai2, Mecp2, Tubb6, Cct8, Map2,

196 Gnb2, Hist1h1c, Hist1h1a, Pcbp3, Snap25, Hist1h1b, Crocc, and Hist1h1e have binding activity, Gnao1, Atp2b1,

197 Pcbp1, Gnai2, Ckb, Prdx6, Gnb2, Ckb, Pcbp3, Atp1a1 and Crocc have catalytic activity, Gnao1 and Gnai2 have

198 signal transducer activity, Tubb6 has structural molecule activity and Slc25a22, Atp2b1, Atp1a1 and Slc6a11 have

199 transporter activity respectively. The DEPs were involved in various cellular components functions like cell part

200 (GO:0044464) (Gnao1, Atp2b1, Gnai2, Mecp2, Tubb6, Cct8, Map2, Prdx6, Snap25 and Slc6a11), macromolecular

201 complex (GO:0032991) (Gnao1, Pcbp1, Gnai2, Cct8, Gnb2, Pcbp3 and Snap25), membrane (GO:0016020) (Gnao1,

202 Atp2b1, Gnai2 and Snap25) and organelle (GO:0043226) (Atp2b1, Mecp2, Tubb6 and Cct8) respectively (Figure

203 2D). DEPs are belonging to different classes of proteins like chaperone (Hsp90ab1 and Cct8), cytoskeletal protein

204 (Tubb6), enzyme modulator (Gnao1, Pcbp1, Gnai2, Gnb2, Pcbp3 and Crocc), hydrolase (Atp2b1, Pcbp1, Gnb2,

205 Pcbp3 and Atp1a1), membrane traffic protein (Snap25), nucleic acid binding (Pcbp1, Hist1h1d, Mecp2, Hist1h1c,

206 Hist1h1a, Pcbp3, Hist1h1b and Hist1h1e), oxidoreductase (Prdx6), transferase (Mecp2 and Ckb), transporter

207 (Atp2b1, Atp1a1 and Slc6a11) and viral protein (Crocc) respectively (Figure 2E). The up and down regulated

208 proteins were involved in various reactome biological pathways like G-protein activation pathway (Gnao1 and

209 Gnai2), Co-operation of PDCL (PhLP1) and TRiC/CCT in G-protein beta folding pathway (Gnao1, Gnai2, Cct8 and

210 Gnb2), Chaperonin-mediated protein folding pathway (Gnao1, Gnai2, Cct8 and Gnb2), Protein folding pathway

211 (Gnao1, Gnai2, Cct8 and Gnb2), Regulation of insulin secretion pathway (Gnai2, Gnb2 and Snap25), mRNA

212 Splicing - Major Pathway (Hnrnpa1, Pcbp1, Hnrnpc and Hnrnpd), Transmission across Chemical Synapses pathway

213 (Gnai2, Gnb2, Snap25 and Slc6a11) respectively (Figure 2F). Table 2 showed the up and down-regulated proteins,

214 gene name, MS/MS count, unique sequence coverage, molecular weight (MW), T-test difference, p-value, various

215 KEGG name, and pathways.

216

217 The Akita vs C57 combination up and down regulated proteins were involved in various molecular functions,

218 biological process, cellular components, protein classes and biological pathways. The up and down regulated

219 proteins were involved in various biological processes such as biological regulation (GO:0065007), cellular

220 component organization or biogenesis (GO:0071840), cellular process (GO:0009987), developmental process

221 (GO:0032502), localization (GO:0051179), metabolic process (GO:0008152), multicellular organismal process

222 (GO:0032501), response to stimulus (GO:0050896) respectively. Nme1, Prdx6, Gm20390, Nme2, Atp1a1, Atp1b2,

223 Atp1a2 proteins are involved in biological regulation (GO:0065007); Epb4.1l3, Lmnb2, Map2, Cfl1 proteins are

224 involved in cellular component organization or biogenesis (GO:0071840); Hnrnpk, Pcbp1, Gnb1, Epb4.1l3, Nme1,

225 Uqcrq, Map2, Prdx6, Gnb2, Gm20390, Atp6v1a, Nme2, Cfl1, Atp1a1, Atp1b2, Atp1a2 proteins are involved in

226 cellular process (GO:0009987); Hnrnpk, Pcbp1, Nme1, Map2, Gm20390, Nme2 proteins are involved in

227 developmental process (GO:0032502); Hnrnpk, Pcbp1, Stxbp1 proteins are involved in localization (GO:0051179);

228 Hnrnpk, Pcbp1, Nme1, Uqcrq, Gm20390, Prdx6, Atp6v1a, Nme2, Atp1a1, Glul, Atp1b2, Atp1a2 proteins are

229 involved in metabolic process (GO:0008152); Hnrnpk, Pcbp1, Gnb1, Map2, Gnb2, Stxbp1 proteins are involved in

230 multicellular organismal process (GO:0032501); Prdx6, Prdx5 proteins are involved in response to stimulus

231 (GO:0050896) respectively (Figure 4B). The up and down regulated proteins are involved in various molecular

232 functions such as antioxidant activity (GO:0016209) (Prdx6 and Prdx5), binding (GO:0005488) (Hnrnpk, Pcbp1,

233 Gnb1, Map2, Gnb2 and Cfl1), catalytic activity (GO:0003824) (Hnrnpk, Pcbp1, Gnb1, Nme1, Uqcrq, Gm20390,

234 Prdx6, Atp6v1a, Nme2, Atp1a1, Glul, Atp1b2, Atp1a2), structural molecule activity (GO:0005198) (Epb4.1l3 and

235 Cfl1) and transporter activity (GO:0005215) (Uqcrq, Atp6v1a, Stxbp1, Atp1a1, Atp1b2, Atp1a2) respectively

236 (Figure 4C). The DEPs are involved in various cellular component processes such as cell part (GO:0044464)

237 (Epb4.1l3, Uqcrq, Map2, Prdx6, Atp6v1a, Cfl1), macromolecular complex (GO:0032991) (Hnrnpk, Pcbp1, Gnb1,

238 Uqcrq, Gnb2 and Atp1b2), membrane (GO:0016020) (Uqcrq, Atp6v1a, Atp1ab2), andorganelle (GO: 0043226)

239 (Epb4.1l3, Atp6v1a, Cfl1) respectively (Figure 4D). The DEPs belong to various protein classes such as

240 cytoskeletal protein (Cfl1), enzyme modulator (Hnrnpk, Pcbp1, Gnb1 and Gnb2), hydrolase (Hnrnpk, Pcbp1, Gnb1

241 and Gnb2, Atp6v1a, Atp1a1 and Atp1a2), ligase (Glul), membrane traffic protein (Stxbp1), nucleic acid binding

242 (Hnrnpk, Pcbp1 and Atp6v1a), oxidoreductase (Prdx6), transporter (Atp6v1a, Stxbp1, Atp1a1, Atp1b2 and Atp1a2)

243 respectively (Figure 4E). We have also analyzed the reactome pathways of up and down regulated proteins, which

244 are Ion transport by P-type ATPases (Atp1a1, Atp1b2 and Atp1a2), Ion channel transport (Atp6v1a, Atp1a1, Atp1b2

245 and Atp1a2), Transmembrane transport of small molecules (Gnb1, Gnb2, Atp6v1a, Atp1a1, Atp1b2 and Atp1a2),

246 Ion homeostasis (Atp1a1, Atp1b2 and Atp1a2), Cardiac conduction (Atp1a1, Atp1b2 and Atp1a2), Regulation of

247 insulin secretion (Gnb1, Gnb2 and Stxbp1), Integration of energy metabolism (Gnb1, Gnb2 and Stxbp1), mRNA

248 Splicing - Major Pathway (Hnrnpk, Hnrnpa1, Pcbp1, Hnrnpc, and Hnrnpd), Processing of Capped Intron-Containing

249 Pre-mRNA (Hnrnpk, Hnrnpa1, Pcbp1, Hnrnpc, and Hnrnpd), Transmission across Chemical Synapses (Gnb1, Gnb2,

250 Stxbp1 and Glul), Neuronal System (Gnb1, Epb4.1l3, Gnb2, Stxbp1 and Glul) and Unclassified pathways (Lmnb2,

251 Hnrnpab, Nefm, Gm20390, Map2, Ntm and Nme2) respectively (Figure 4F). Table 3 showed the up and down

252 regulated proteins, gene name, MS/MS count, unique sequence coverage, molecular weight (MW), T-test difference,

253 p-value, various KEGG name and pathways.

254 The PlGF-/- vs. C57 combination up and down-regulated proteins were involved in various molecular functions,

255 biological process, cellular components, protein classes and biological pathways. The DEPs are involved various

256 biological processes such as biological regulation (GO:0065007), cellular component organization or biogenesis

257 (GO:0071840), cellular process (GO:0009987), developmental process (GO:0032502), localization (GO:0051179),

258 locomotion (GO:0040011), metabolic process (GO:0008152), multicellular organismal process (GO:0032501),

259 response to stimulus (GO:0050896) respectively. Napb, Rac3, Gnat1, Gnat3, Rhobtb3, Rac2, Rac1, Gnat2, Atp1b2

260 proteins are involved in biological regulation (GO:0065007); Napb, Rac3, Tubb3, Rhobtb3, Epb4.1l3, Rac1,

261 Lmnb2, Tpm1, Tpm2 proteins are involved in cellular component organization or biogenesis (GO:0071840); Napb,

262 Rac3, Gnat1, Gnb1, Gnat3, Tubb3, Ywhab, Rhobtb3, Epb4.1l3, Rac2, Gnb4, Gngt1, Tpm1, Rac1, Gnb2, Rac2,

263 Gnat2, Tpm2, Nono, Crocc, Atp1b2, Aco2, Ywhag are involved in cellular process (GO:0009987); Rac1 and Rac2

264 are involved in developmental process (GO:0032502); Napb and Rac3 are involved in localization (GO:0051179);

265 Rac3 is also involved in locomotion (GO:0040011); Srsf9, Rac3, Nono, Rhobtb3, Rac3, Gnat3, Aco2, Atp1b2,

266 Cirbp, Gnat1, Gnat2, Rac1, Rac2, Srsf1 are involved in metabolic process (GO:0008152); Gnat3, Gnb1, Gnb2,

267 Gnb4, Rac1, Rac2, Napb, Tpm1, Tpm2 are involved in multicellular organismal process (GO:0032501), and Gnat1,

268 Gnat2, Gnat3, Rac3, Rac2, Rac1, Rhobtb3 are involved in response to stimulus (GO:0050896) respectively (Figure

269 5B). The DEPs are involved in various molecular functions such as binding (GO:0005488) (Aco2, Cirbp, Crocc,

270 Gnat1, Gnat2, Gnat3, Gnb1, Gnb2, Gnb4, Gngt1, Napb, Nono, Rac1, Rac2, Rac3, Rhobtb3, Srsf1, Srsf9, Tmp1,

271 Tmp2 and Tubb3), catalytic activity (GO:0003824) (Aco2, Cirbp, Crocc, Gnat1, Gnat2, Gnat3, Gnb1, Gnb2, Gnb4,

272 Gngt1, Napb, Nono, Rac1, Rac2, Rac3, Rhobtb3, Srsf1, Srsf9), signal transducer activity (GO:0004871) (Gnat1,

273 Gnat2, Gnat3, Rac1, Rac2, Rac3), structural molecule activity (GO:0005198) (Epb4.1l3 and Tubb3), transporter

274 activity (GO:0005215) (Atp1b2) respectively (Figure 5C). The DEPs are involved in cellular component process

275 such as cell part (GO:0044464) (Aco2, Epb4.1l3, Gnat1, Gnat2, Gnat3, Napb, Rac3, Ndufa8, Nono, Rhobtb3,

276 Tmp1, Tmp2, and Tubb3), macromolecular complex (GO:0032991) (Atp1b2, Gnat1, Gnat2, Gnat3, Gnb1, Gnb2,

277 Gnb4, Gngt1, Napb, Ndufa8, Tpm1, and Tpm2), membrane (GO:0016020) (Atp1b2, Gnat1, Gnat2, Gnat3, Napb,

278 Ndufa8, Rac3 and Rhobtb3), organelle (GO:0043226) (Aco2, Epb4.1l3, Napb, Rac3, Nono, Tpm1, Tpm2, Rhobtb3

279 and Tubb3), and synapse (GO:0045202) (Napb) respectively (Figure 5D). The DEPs are classified into various

280 protein class such as chaperone (Ywhab and Ywhag), cytoskeletal protein (Tubb3, Tpm1 and Tpm2), enzyme

281 modulator (Rac3, Gnat1, Gnb1, Gnat3, Rhobtb3, Rac2, Gnb4, Gngt1, Rac1, Gnb2, Gnat2 and Crocc), hydrolase

282 (Gnb1, Gnb2, Gnb4), lyase (Aco2), membrane traffic protein (Napb), nucleic acid binding (Cirbp, Srsf9, Srsf1 and

283 Nono), oxidoreductase (Ndufa8), transcription factor (Rac3), transferase (Rac3), transporter (Atp1b2), and viral

284 protein (Crocc) respectively (Figure 5E). We have also analyzed the reactome pathways of up and down regulated

285 proteins, which are Activation of the phototransduction cascade (Gnat1, Gnb1 and Gnat2), Signal Transduction

286 (Rac3, Gnat1, Gnb1, Gnat3, Tubb3, Ywhab, Rac2, Gnb4, Gngt1, Rac1, Gnb2, Gnat2, Gfap and Ywhag), G-protein

287 activation (Gnat1, Gnb1, Gnat3 and Gnat2), Opioid Signalling (Gnat1, Gnb1, Gnat3 and Gnat2), Signalling by

288 GPCR (Gnat1, Gnb1, Gnat3, Ywhab, Rac2, Gnb4, Gngt1, Gnb2, Rac1 and Gnat2), ADP signalling through P2Y

289 purinoceptor-12 (Gnb1 and Gnat3), Platelet activation, signalling and aggregation (Gnb1, Gnat3, Rac2 and Rac1),

290 Hemostasis (Gnb1, Gnat3, Rac2 and Atp1b2), Cooperation of PDCL (PhLP1) and TRiC/CCT in G-protein beta

291 folding (Gnat1, Gnb1, Gnat3, Gnb4, Gnb2 and Gnat2), Chaperonin-mediated protein folding (Gnat1, Gnb1, Gnat3,

292 Gnb4, Gnb2 and Gnat2), Metabolism of proteins (Gnat1, Gnb1, Gnat3, Eif4a1, Gnb4, Eif4a2, Gnb2, Hnrnpc and

293 Gnat2), Activation of BAD and translocation to mitochondria (Ywhab and Ywhag), Intrinsic Pathway for Apoptosis

294 (Ywhab and Ywhag) Adrenaline, noradrenaline inhibits insulin secretion (Gnb1, Gnb2, Gnb4 and Gngt1),

295 Regulation of insulin secretion (Gnb1, Vamp2, Gnb2, Gnb4 and Gngt1), Metabolism (Atp5o, Gnb1, Vamp2,

296 Ndufa8, Atp5f1, Gnb4, Gngt1, Gnb2, Aco2 and Hba), RHO GTPase Effectors (Ywhab, Rac2, Rac1 and Ywhag),

297 Signaling by Rho GTPases (Ywhab, Rac2, Rac1 and Ywhag), Glucagon-type ligand receptors (Gnb1, Gnb2, Gnb4

298 and Gngt1), Activation of G protein gated Potassium channels (Gnb1, Gnb2, and Gngt1), Transmission across

299 Chemical Synapses (Gnb1, Gnat3, Vamp2, Gngt1 and Gnb2), Glucagon-like Peptide-1 (GLP1) regulates insulin

300 secretion (Gnb1, Gnb2, Gnb4 and Gngt1), Inhibition of voltage gated Ca2+ channels via Gbeta/gamma subunits

301 (Gnb1, Gnb2, and Gngt1), Neuronal System (Gnb1, Gnat3, Vamp2, Epb4.1l3, Gngt1 and Gnb2), Inactivation,

302 recovery and regulation of the phototransduction cascade (Gnat1 and Gngt1), Beta-catenin independent WNT

303 signalling (Gnb1, Rac1 and Gnat2), Rho GTPase cycle (Rac1, Rac2, and Rac3) and un classified pathway (Cirbp,

304 Napb, Vamp1, Lmnb2, Nefm, Nono, Crocc and Rbp3) respectively (Figure 5F). Table 4 showed the up and down-

305 regulated proteins, gene name, MS/MS count, unique sequence coverage, molecular weight (MW), T-test difference,

306 p-value, various KEGG name, and pathways.

307 The Akita.PlGF-/- vs. PlGF-/- combination up and down-regulated proteins are involved in various molecular

308 functions, biological processes, cellular components, protein classes and biological pathways. The DEPs are

309 involved in various biological process such as biological regulation (GO:0065007), cellular component organization

310 or biogenesis (GO:0071840), cellular process (GO:0009987), developmental process (GO:0032502), localization

311 (GO:0051179), metabolic process (GO:0008152), multicellular organismal process (GO:0032501), and response to

312 stimulus (GO:0050896) respectively. Map1b and Hnrnpm are involved in biological regulation (GO:0065007),

313 Tubb2b, Tubb3, Tubb6, Cct8, Map1b, and Tubb5 are involved in cellular component organization or biogenesis

314 (GO:0071840), Hspa8, Tubb2b, Tubb3, Tubb6, Cox5b, Cct8, Map1b, Hnrnpm, Tubb5, and Pura proteins are

315 involved in cellar process (GO:0009987), Map1b is involved in developmental process (GO:0032502), Hspa8,

316 Vdac2 and Cox5b are involved in localization (GO:0051179) biological process, Cirbp, Hspa8, Hsp90ab1, Cox5b,

317 Cct8, Erh, Hnrnpm and Pura protein are involved in metabolic process (GO:0008152), Map1b is involved in

318 multicellular organismal process (GO:0032501), and Hspa8, Hsp90ab1, Mecp2, Hnrnpm are involved in response to

319 stimulus (GO:0050896) respectively (Figure 6B). The DEPs were involved in various molecular functions such as

320 binding (GO:0005488) (Cirbp, Hspa8, Tubb2b, Tubb3, Mecp2, Tubb6, Cct8, Map1b, Erh, Hnrnpm, Tubb5 and

321 Pura), catalytic activity (GO:0003824) (Cirbp, Hspa8, and Cox5b), structural molecule activity (GO:0005198)

322 (Tubb2b, Tubb3, Tubb6, and Tubb5), and transporter activity (GO:0005215) (Vdac2 and Cox5b) respectively

323 (Figure 6C). The DEPs were also involved in various cellular component functions such as (Figure 6D). The DEPs

324 are belonged to various protein classes such as cell adhesion molecule (Rs1), chaperone (Hsp90ab1, Cct8),

325 cytoskeletal protein (Tubb2b, Tubb3, Tubb6, Map1b and Tubb5), hydrolase (Rs1), nucleic acid binding (Cirbp,

326 Mecp2, Hnrnpm, Pura), oxidoreductase (Ndufa8, Cox5b), receptor (Rs1), signalling molecule (Rs1), transcription

327 factor (Erh, Pura), transferase (Mecp2) and transporter (Vdac2) respectively (Figure 6E). The DEPs are also

328 involved in various reactome pathways such as Attenuation phase (Hspa8 and Hsp90ab1), HSF1-dependent

329 transactivation (Hspa8 and Hsp90ab1), Intraflagellar transport (Tubb2b, Tubb3, and Tubb6), Assembly of the

330 primary cilium (Tubb2b, Tubb3, Tubb6, and Tubb5), Organelle biogenesis and maintenance (Tubb2b, Tubb3,

331 Tubb6, and Tubb5), Translocation of GLUT4 to the plasma membrane (Tubb2b, Tubb3, and Tubb6), Hedgehog 'off'

332 state (Tubb2b, Tubb3, and Tubb6), and Hedgehog 'on' state (Tubb2b, Tubb3, and Tubb6) respectively (Figure 6F).

333 Table 5 showed the up and down-regulated proteins, gene name, MS/MS count, unique sequence coverage,

334 molecular weight (MW), T-test difference, p-value, various KEGG name, and pathways.

335

336 Protein-protein network analysis

337 The protein-protein network analysis is a wide-ranging approach to know the annotation of desire proteome (Kumari

338 et al., 2015). The functional network protein study will be helpful for drug discovery, to understand metabolic

339 pathways and to predict or develop genotype-phenotype associations (Wang and Moult, 2001; Wang et al., 2009).

340 We have performed the protein-protein network analysis for all DEPs using STRING 10 database (https://string-

341 db.org/). The DEPs of Akita.PlGF-/- vs. Akita have uploaded to STRING network set the mouse database. They have

342 total 30 nodes, 26 edges, cluster coefficient 0.497, average node degree:1.73, PPI enrichment p-value:1.17e-05. In

343 KEGG pathway network, Gnb2, Gnai2, and Gnao1 are involved in Retrograde endocannabinoid signaling pathway

344 (ko04723), Glutamatergic synapse pathway (ko04724), Cholinergic synapse pathway (ko04725), GABAergic

345 synapse pathway (ko04727), Dopaminergic synapse pathway (ko04728), Serotonergic synapse pathway (ko04726).

346 Pcb1, Hnrnpc, and Hnrnpa1 are involved in Spliceosome pathway (ko03040), Snap25 and Atp6v1e1 are involved in

347 Synaptic vesicle cycle pathway (ko04721). Figure 3G highlighted the up and down-regulated proteins are involved

348 in complex protein assembly (yellow) in biological process (GO:0006461), binding (blue) function of molecular

349 function (GO:0005488), membrane-bounded organelle (red) proteins in cellular component (GO:0043227). The

350 Akita vs. C57 up, and down-regulated proteins have total 28 nodes, 27 edges, cluster co-efficient 0.539, average

351 node degree:1.73, PPI enrichment p-value: 2.62e-05. Figure 4G highlighted the up and down-regulated proteins

352 involved in nervous system development (GO:0007399) (red), response to glucose (GO:0009749) (blue) of

353 biological process, catalytic activity (GO:0003824) (light green), hydrolase activity (GO:0016787) (pink), of

354 molecular functions, Insulin secretion (ko04911) (yellow), Pancreatic secretion (ko04972) (cyan), Oxidative

355 phosphorylation (ko00190) (thick green) of KEGG pathways. The PlGF-/- vs. C57 up, and down-regulated proteins

356 have total 40 nodes, 87 edges, cluster co-efficient 0.692, average node degree:4.35, PPI enrichment p-value: 9.93e-

357 10. Figure 5G highlighted the up and down regulated proteins involved in nervous system development

358 (GO:0007399) (red), eye photoreceptor cell differentiation (GO:0001754) (blue) of biological process, hydrolase

359 activity (GO:0016787) (light green), catalytic activity (GO:0003824) (yellow) of molecular functions, photoreceptor

360 inner segment (GO:0001917) (pink), photoreceptor outer segment (GO:0001750) (thick green) of cellular

361 components, Ras signalling pathway (ko04014) (cyan), VEGF signalling pathway (ko04370) (orange), Oxidative

362 phosphorylation (ko00190) (magenta) of KEGG pathways. The Akita.PlGF-/- vs PlGF-/- up and down-regulated

363 proteins have total 19 nodes, 14 edges, cluster co-efficient 0.511, average node degree:1.47, PPI enrichment p-value:

364 0.0013. Figure 6G highlighted the up and down-regulated proteins involved in protein folding (GO:0006457) (blue)

365 of biological process, binding (GO:0005488) (red) of molecular function, intracellular organelle part (GO:0044446)

366 (green) of cellular component, Gap junction (ko04540) (green) of KEGG pathway. The three and four combinations

367 of up and down-regulated proteins are represented as a Venn diagram (Figure 7A, B), Akita vs. C57 have 30 up,

368 and down-regulated proteins, 20 (26.3%) proteins are specific, 9 (11.8%) proteins are shared with PlGF-/- vs. C57,

369 only one (1.3%) protein is shared with other two groups. Akita. PlGF-/- vs. Akita have 15 specific proteins; 3

370 proteins are shared with PlGF-/- vs. C57, and one protein is shared with other two groups.

371

372 3.0 Discussion

373 To elucidate molecular mechanisms that PlGF mediates in early complications in DR, we examine retinal proteome

374 of the four mouse strains Akita.PlGF-/-, Akita, C57, and PlGF-/- using label-free mass spec quantification. This

375 approach provides protein information for all retinal cell types during mass spectral analysis. Using normalization

376 by the Z-score method, correlation by Pearson correlation coefficient, we have identified differentially expressed

377 proteins in Akita.PlGF-/- vs Akita (31 proteins), Akita vs C57 (26 proteins), PlGF-/- vs C57 (31 proteins),

378 Akita.PlGF-/- vs. PlGF-/- (19 proteins) combinations, which may be involved in the protective retinal phenotype in

379 diabetes retinopathy conditions.

380 In Akita.PlGF-/-, Map2, Prdx6, Tubb6, Crocc, Hsp90ab1 and Ckb were up-regulated and Gnb2, Snap25, Pcbp1,

381 Atp1a1, Pcbp3, Gnao1, Atp2b1, Gnai2, Vim, Slc25a22, Hnrnpa1, Mecp2 etc., are downregulated when compared to

382 Akita group (Table 2). Functional classification and pathway analysis results revealed that Map2 gene encodes

383 microtubule-associated proteins, which are involved in the binding activity of molecular function, cellular

384 component organization or biogenesis, and cellular process. These proteins are thought to be comprised in

385 microtubule assembly, which is a crucial step in neuritogenesis. In rat and mouse, Map2 are neuron-specific

386 cytoskeletal proteins that are augmented in dendrites, associating a role in determining and stabilizing dendritic

387 shape throughout neuron development (Lim & Halpain, 2000). Increased productions of IL-1b, IL-6, or TNF-a

388 decrease the Map2 expression and lead to neuronal cell death. However, the role of the Map2 is not limited to

389 neuronal differentiation or merely neuron cell loss cell marker. Map2 serves to stabilize microtubules (MT) growth

390 by crosslinking microtubules with intermediate filaments. Map2 upregulation increases neuronal cell survival and

391 differentiation. Furthermore Map2 also has a binding domain for the regulatory subunit II of cAMP-dependent

392 protein kinase (PKA). It is known that in diabetes increased oxidative stress and low energy levels contribute to

393 neuron microtubule meshwork degeneration that eventually leads to neuron cell death (Harada et al, 2002;

394 Iriuchijima et al, 2005; Teng et al, 2001). Our results revealed that Map2 protein is up-regulated in Akita.PlGF-/-

395 compared to Akita mice retinas. Peroxiredoxin-6 (Prdx6) protein is up-regulated in the Akita.PlGF-/- compare to

396 Akita and belongs to oxidoreductase class, the antioxidant activity of molecular function, involved in cellular and

397 metabolic process. Prdx isoforms (Prdx) are expressed in the majority of mammalian cells, which differ in structure,

398 catalytic mechanisms and subcellular compartmentalization (Hanschmann et al, 2013). In the retina, Prdx6 was

399 expressed solely by Müller cells and astrocytes. Astrocytes and Müller cells play crucial structural and functional

400 roles in the maintenance of barrier function, uniquely within the retina; decreased Prdx6 is of obvious relevance to

401 any disease where the blood-retinal barrier is compromised, such as diabetic retinopathy, exudative age-related

402 macular degeneration and arterial and venous occlusions (Chidlow et al, 2016).

403 Tubulin beta6 class V (Tubb6) is a cytoskeletal protein, has binding activity, the structural molecular activity of

404 molecular functions, cellular component organization or biogenesis, cellular process of biological process. Tubb6

405 allows to modify the organization of muscle microtubules, regardless of the presence or absence of dystrophin, for

406 the guidance and proper organization of microtubules (Oddoux et al, 2013). Crocc (Rootletin) is a class of enzyme

407 modulator, has catalytic activity and is involved in the cellular process. The rootlet aids to anchor the cilium to the

408 cell and functions as a channel for proteins intended for the outer segment of rod photoreceptors (Yang et al, 2005).

409 Hsp90ab1 is a chaperone class of protein that binds to other proteins, thereby stabilizing them in an ATP-dependent

410 manner (Taipale et al, 2010). Creatine kinase B type (Ckb) is a vital enzyme for energy metabolism. It catalyzes the

411 conversion of creatine, consuming adenosine triphosphate (ATP), phosphocreatine and adenosine diphosphate

412 (ADP), in the visual cycle. Ckb is critical for providing energy for the visual cycle in photoreceptors (Zhu et al,

413 2013). Guanine Nucleotide Binding proteins (Gnb1, Gnb2, Gnb4, Gnai2, Gnao1) are called as G-proteins, which

414 play a key role in insulin signaling pathways. Incessant activation of G-proteins (polymorphism) results in insulin

415 resistance and ultimately increases hepatic glucose output. Gnb gene dimorphism leads to increased cardiac

416 potassium channel activity and increased α-adrenoceptor–mediated vasoconstriction thus resulting in the

417 development and progression of hypertension, obesity and insulin resistance in humans (Benjafield et al, 2001; Poch

418 et al, 2002).

419 Synaptosomal-associated protein 25 (Snap25) is a component of the trans-SNARE complex, which is proposed to

420 account for the specific membranes fusion and to directly execute fusion by forming a tight complex that brings the

421 synaptic vesicle and plasma membranes together. Zuheng et al. reported that a lowering of Snap25 can be associated

422 with an enhancement of insulin secretion (Ma et al, 2005), but both stimulatory and inhibitory influences by

423 physical interaction of Snap25 with L-Type Ca2+ channels in β cells (Ji et al, 2002). Our mass spectral results are

424 also correlated to Ma et al. (Ma et al, 2005). Snap25 protein is down-regulation in Akita.PlGF-/- group compared to

425 Akita group. Ji et al. supposed to L-Type Ca2+ channels interaction with Snap25 and inhibitory and stimulation of

426 insulin, in the same way, our results also supported because of down-regulation of plasma membrane calcium-

427 transporting ATPase1 (Atp2b1 protein) (Ji et al, 2002).

428 In Akita vs C57 comparison, we have found various classes of up and down-regulated proteins which are involved

429 in the different biological processes, molecular functions, cellular components and Reactome pathways. In

430 particular, we have highlighted Ion transport by P-type ATPases, regulation of insulin secretion and processing of

431 capped intron-containing pre-mRNA. Atp1a1, Atp1b2, and Atp1a2 are up-regulated in Akita compared to the C57

432 group, which are involved in ROS pathways and lead to obesity, insulin resistance, and metabolic syndrome (Sodhi

433 et al, 2015). Gnb1, Gnb2, and Stxbp1 are involved in regulation of insulin secretion pathway in up-regulation in

434 Akita compared to the C57 group, but these are insulin resistance in up-regulation condition. Andersson et al

435 (Andersson et al, 2012) supported that reduced expression of exocytotic genes (Stxbp1, Snap25, Vamp2) contributes

436 to impaired insulin secretion, and suggested that decreased expression of these genes as part of diabetes condition.

437 Hnrnpk, Hnrnpa1, Hnrnpc, and Hnrnpd are involved in the processing of capped intron-containing pre-mRNA

438 pathway. He and Smith (He & Smith, 2009) suggested that Hnrnp1 and Hnrnp2 are involved in multiple cellular

439 processes, including DNA replication and transcription and mRNA export, as well as pre-mRNA splicing

440 regulation. The upregulation of Hnrnp (Heterogeneous nuclear ribonucleoprotein) proteins on alternative splicing

441 could lead to significant changes in proteomic diversity in a cell population. In our results showed that these Hnrnp

442 proteins are upregulated in Akita compared to C57 but downregulat in Akita.PlGF-/- group.

443 In PlGF-/- vs C57 group we have identified 31 up and down-regulated proteins, which belong to various classes

444 (Figure 6), are involved in different biological process pathways, molecular functions, cellular components and

445 Reactome pathways. Specifically, Gnb1, Gnb2, Gngt1, Gnb4, Vamp2 proteins, which are involved in diabetes

446 directly or indirectly, also up-regulated in PlGF-/- group compared with the C57 group but when compared to Akita

447 are down-regulated.

448 We have also found up-regulated protein (Ndufa8, Atp5o, Atp5f1) involved in respiratory electron transport, ATP

449 synthesis by chemiosmotic coupling, and heat production by uncoupling proteins. NADH dehydrogenase

450 [ubiquinone] 1 alpha subcomplex subunit 8 (Ndufa8) was mitochondrial intermembrane space assembly pathway.

451 This minor pathway depends on redox-regulated folding events that stabilize, trap and finally import into the

452 mitochondrion the C-X9-C domain-carrying proteins (Schmidt et al, 2010). Eukaryotic translation initiation factor

453 4α2 (Eif4a2), Eif4a1 are deadenylation of mRNA, upstream of the Eif4a2 gene (Cheyssac et al, 2006), Eif4a1

454 showed a significant association with type 2 diabetes but which down-regulation in PlGF-/- group compared to the

455 C57 group.

456 We have also compared Akita to PlGF-/-, significantly Hexokinase (Hk1), Stxbp1, Hnrnpc, Vamp2, Vamp3, Atp1b1,

457 Pcbp1, Pcbp2, Slc6a11, and Nono are up-regulated in Akita, which is mostly involved in diabetes.

458 Rcvrn (Recoverin), Sag, Epb4, Crocc, Cplx4, Gnat1, and Gngt1 are down-regulated in Akita group. Recoverin

459 protein is a neuronal calcium-binding protein that is primarily detected in the photoreceptor cells of the eye. It plays

460 a crucial role in the inhibition of rhodopsin kinase, a molecule which regulates the phosphorylation of rhodopsin. A

461 reduction in this inhibition helps regulate sensory adaptation in the retina, since the light-dependent channel closure

462 in photoreceptors causes calcium levels to decrease, which relieves the inhibition of rhodopsin kinase by calcium-

463 bound recoverin (Chen et al, 1995).

464 Our results also correlated the earlier studies because the recoverin is known to be down regulated in diabetic

465 condition. Akita.PlGF-/- vs. PlGF-/- combination, significantly Tubb3, Tubb6, Tubb2b, Tubb5, Hspa8, Hsp90ab1

466 were up-regulated while Atp5o, Map1b, Hnrnpm, Vdac2, Pura, Cox5b Ndufa8 and Erh proteins are down-regulated

467 (Table 4) in Akita.PlGF-/- compared to PlGF-/-.

468

469 4.0 Conclusion

470 Here we have analyzed the proteomics changes in associated with PlGF ablation in diabetic and healthy condition.

471 Identifying the number of the proteins that can contribute to better understanding molecular mechanisms of PlGF

472 function its target and off-target effects.

473 Reduced insulin resistance of Akita.PlGF-/- are potentially associated with P-type ATPases and GNB group proteins

474 involve in cell metabolism that allow retinal cells to better utilize the glucose reducing the metabolic stress and ROS

475 production. Particularly major neuron survival factor MAP2 and antioxidant defence protein Prdx6 are upregulated

476 in PlGF-/- and Akita.PlGF-/- conditions. Increased expression of Tubb and Hsp protein groups in Akita.PlGF-/-

477 animals indicated increase cell chaperon activity and integrity that potentially increases retinal cell survival. We

478 intend this proteome wide analysis will encourage further research into identified pathways of PlGF effects in DR.

479

480 5.0 Materials and Methods

481 5.1 Mouse Strains

482 The use of animals was in compliance with the Association for Research in Vision and Ophthalmology (ARVO)

483 Statement for the Use of Animals in Ophthalmic and Vision Research and approved by the Institutional Animal

484 Care and Use Committee of The Johns Hopkins University (Protocol number: M016M480). Mice were housed at

485 the special pathogen-free (SPF) Cancer Research Building Animal Facilities at Johns Hopkins Hospital. Mouse

486 strains

487 C57BL/6-Ins2/J.PlGF-/-, C57BL/6-PlGF-/-, C57BL/6-Ins2/J, and C57BL/6J (C57) were generated

488 and maintained as described previously. (Huang et al, 2015b) Briefly, Akita mice were crossed with PlGF−/− mice

489 (Van de Veire et al, 2010) in a C57BL/6J background for two generations to give birth to the progeny with the

490 genotype of Akita.PlGF−/−. The breeding program graphical abstract is presented in the Figure 1A.

491 492 5.3 Verification of diabetic conditions

493 Diabetic phenotype and genotype were confirmed 4.5 weeks after birth by blood glucose >250 mg/dL (One-Touch

494 Lifescan meter; Lifescan, Inc., Milpitas, CA) in a drop of blood from a tail puncture Diabetes development were not

495 maintained with insulin. Figure 1B The levels of blood glucose were converted to glycated hemoglobin (HbA1c)

496 using an online tool (https://www.accu-chek.com/us/glucose-monitoring/a1c-calculator.html). Final blood glucose

497 concentration and body weight were measured before sacrifice.

498 499 5.4 Retinal samples collection

500 Mice were euthanized by ketamine overdose (100 mg/kg), intraperitoneal injection. Mouse eyes were enucleated,

501 and retinas were carefully isolated free of the cornea, lens, vitreous humor and retinal pigment epithelium (RPE) on

502 ice in cold phosphate-buffered saline (PBS, 0.1M, pH=7.2) under the dissection microscope Stemi 2000-C (Zeiss).

503 Retinas were frozen with liquid nitrogen and then stored at -80◦C until further use. (Figure 1C)

504

505 5.5 Protein extraction and peptide preparation

506 Retinas were homogenized using RIPA buffer (R0278-50ML; Sigma) supplemented with Protease and Phosphates

507 Inhibitor cocktail (Cell Signalling Technology). Retinal homogenates were centrifuged (12,000g, 4°C, and 10min)

508 to remove tissue debris. The protein concentration of the supernatant was determined using Pierce BCA Protein

509 Assay Kit (Thermo Fisher Scientific, Cat#: 23225); 50-microgram of total protein from each sample was used for

510 peptide preparation. DTT reduced proteins and alkylated by iodoacetamide (10mM, in 50mM Tris-HCl, pH8.0) at

511 room temperature for 1h. The protein samples were then digested with trypsin/LysC (Promega, Cat#: V5073) in

512 25mM ammonium bicarbonate solution at 37°C for 12 h. The resultant peptides were washed and desalted with the

513 desalting column (Pierce Spin-Tip) and eluted with 50µl 5% acetonitrile + 0.4% trifluoroacetic acid. The peptides

514 mixtures were run through the ziptip C18 (Millipore) and then dried with SpeedVac concentrator. (Figure 1D)

515

516 5.3 Mass spectrometry

517 The LC/MS/MS analysis of samples was carried out using a Thermo Scientific Q-Exactive hybrid Quadrupole-

518 Orbitrap Mass Spectrometer and a Thermo Dionex UltiMate 3000 RSLCnano System (Poolchon Scientific,

519 Frederick, MD, USA). Peptide mixtures from each sample were loaded onto a peptide trap cartridge at a flow rate of

520 5 μL/min. The trapped peptides were eluted onto a reversed-phase PicoFrit column (New Objective, Woburn, MA)

521 using a linear gradient of acetonitrile (3-36%) in 0.1% formic acid. The elution duration was 120 min at a flow rate

522 of 0.3 μl/min. Eluted peptides from the PicoFrit column were ionized and sprayed into the mass spectrometer, using

523 a Nanospray Flex Ion Source ES071 (Thermo) under the following settings: spray voltage, 1.6 kV, Capillary

524 temperature, 250°C. 18

525

526 5.4 Label-free proteomics data analysis:

527 The mass spectral (MS) raw data were analyzed with MaxQuant computational proteomics platform (version

528 1.6.1.0) and its built-in Andromeda search engine (Cox & Mann, 2008). The LTQ-Orbitrap peptides were identified

529 (main search peptide search = 4.5 ppm and 20 ppm for first search peptide tolerance, respectively) from the MS/MS

530 spectra searched against Mus musculus UniProtKB (83,598 entries) target database (Perumal et al, 2014) using

531 Andromeda search engine. (Cox et al, 2011) This target database was also shared with the common contaminants

532 and concatenated with the reversed versions of all sequences. In group-specific parameters, enzyme-specific was set

533 to trypsin, and two missed cleavages were allowable; variable modifications were selected as fixed modification

534 whereas carbamidomethylation (C) and oxidization (M) were set as fixed and variable modifications respectively

535 were selected as variable modifications and 5 set to a maximum number of modifications per peptide. The type was

536 set the standard; multiplicity was set to 1 to account for the label-free state, and Label-free quantification (LFQ) was

537 set LFQ, 2 set for minimum ration count (Cox & Mann, 2008). The FDR (false discovery rate) was set to 0.01 for

538 proteins, peptides and other parameters were set to default. Label minimum ratio count was set to 2, peptides for

539 quantification was set to unique and razor and re-quantify to allow identification and quantification of proteins in

540 groups for LFQ analysis.

541 First, MaxQuant corrects for systematic inaccuracies of measured peptide masses and corresponding retention times

542 of extracted peptides from the raw data. Then for peptide identification, mass and intensity of the peptide peaks in

543 mass spectrometry (MS) spectrum are detected and assembled into three-dimensional (3D) peak hills over the m/z

544 retention time plane, which is filtered by applying graph theory algorithms to identify isotope patterns. High mass

545 accuracy is achieved by weighted averaging and through mass recalibration by subtracting the determined

546 systematic mass error from the measured mass of each MS isotope pattern. Peptide and fragment masses (in case of

547 an MS/MS spectrum) are searched in an organism-specific sequence database and are then scored by a probability-

548 based approach termed peptide score. For limiting a certain number of peak matches by chance, a target-decoy-

549 based false discovery rate (FDR) approach is utilized. The FDR is determined using statistical methods that account

550 for multiple hypotheses testing. Also, the organism-specific database search includes not only the target sequences

551 but also their reverse counterparts and contaminants, which helps to determine a statistical cut-off for acceptable

552 spectral matches. The assembly of peptide hits into protein hits to identify proteins is the next step, in which each

553 identified peptide of a protein contributes to the overall identification accuracy. Also, an FDR-controlled algorithm

554 called matching between runs is incorporated, which enables the MS/MS free identification of MS features in the

555 complete data set for every single measurement, leading to an increase in the number of quantified proteins per

556 sample (Cox & Mann, 2008).

557

558 5.5 Perseus analysis pipeline

559 The MaxQuant software generated output file “proteingroups.txt” was utilized for Pearson correlation, clustering

560 and statistical analysis using Perseus software version 1.6.1.1 (Tyanova et al, 2016). Unverified hierarchical

561 clustering of the LFQ values was carried out based on Euclidean distances on the Z-scored among mean values. For

562 statistical analysis, two-samples t-test-based statistics with P < 0.05 was applied on Log2 transformed LFQ values

563 and the minimum number of values “in at least one group” is 3 to assert proteins regulation as significant for the

564 specific groups (Cox et al, 2014; Perumal et al, 2015).

565

566 5.6 Functional classification and pathway analysis

567 Proteins determined to be differentially expressed as illustrated based on the data in our LFQ experiments were

568 tabularize in Excel and their gene names were used for functional annotation and pathways analysis. First, DAVID

569 tool (v6.7) (Huang da et al, 2009) (http://david.abcc.ncifcrf.gov/home.jsp) was used for interpreting the GOBP terms

570 of the differentially-expressed proteins. The protein list was uploaded into DAVID and searched for enrichment for

571 GOBP term, and the results were filtered based on threshold count ≥ 2 and P values < 0.05.

572

573 5.7 Protein-protein network analysis:

574 We have also performed the functional enrichment and interaction network analysis using STRING 10.5 database

575 (Szklarczyk et al, 2011) on differentially expressed proteins. SPRING tool classified the proteins according to the

576 Gene Ontology (GO) categories such as biological process (BP), molecular function (MF), Cellular Component

577 (CC) and KEGG (Kyoto Encyclopedia of Genes and Genomes database) pathways. (Kanehisa et al, 2017) The tool

578 Venny 2.1 (http://bioinfogp.cnb. csic.es/tools/venny/) was used to generate Venn diagrams.

579

580 5.9 Statistical analysis

581 All values were expressed as the mean ± standard deviation (SD) for the respective groups. Statistical analyses were

582 performed with GraphPad Prism software (https://www.graphpad.com/scientific-software/prism/). The Student’s t-

583 test, one-way ANOVA test with Tukey multiple comparisons, were used. P value less than 0.05 were considered

584 significant.

585 586 5.10 Accession numbers 20

587 The accession number for the raw mass spectrometry data reported in this study is ProteomeXchange Consortium:

588 (Mass Spec Raw data will be uploaded during the revision process)

589

590 6.0 References

591 A Study to Evaluate the Safety and Efficacy of THR-317 for the Treatment of Diabetic Macular Oedema (DME). 592 https://ClinicalTrials.gov/show/NCT03071068. 593 594 A Study to Evaluate THR-317 and Ranibizumab Combination Treatment for Diabetic Macular Oedema (DME). 595 https://ClinicalTrials.gov/show/NCT03499223. 596 597 Andersson SA, Olsson AH, Esguerra JL, Heimann E, Ladenvall C, Edlund A, Salehi A, Taneera J, Degerman E, 598 Groop L, Ling C, Eliasson L (2012) Reduced insulin secretion correlates with decreased expression of exocytotic 599 genes in pancreatic islets from patients with type 2 diabetes. Mol Cell Endocrinol 364: 36-45 600 601 Barber AJ (2015) Diabetic retinopathy: recent advances towards understanding neurodegeneration and vision loss. 602 Sci China Life Sci 58: 541-549 603 604 Barber AJ, Antonetti DA, Kern TS, Reiter CE, Soans RS, Krady JK, Levison SW, Gardner TW, Bronson SK (2005) 605 The Ins2Akita mouse as a model of early retinal complications in diabetes. Invest Ophthalmol Vis Sci 46: 2210-2218 606 607 Benjafield AV, Lin RC, Dalziel B, Gosby AK, Caterson ID, Morris BJ (2001) G-protein beta3 subunit gene splice 608 variant in obesity and overweight. Int J Obes Relat Metab Disord 25: 777-780 609 610 Carmeliet P, Jain RK (2011) Molecular mechanisms and clinical applications of angiogenesis. Nature 473: 298-307 611 612 Chen CK, Inglese J, Lefkowitz RJ, Hurley JB (1995) Ca(2+)-dependent interaction of recoverin with rhodopsin 613 kinase. J Biol Chem 270: 18060-18066 614 615 Chen L, Tao Y, Jiang Y (2015) Apelin activates the expression of inflammatory cytokines in microglial BV2 cells 616 via PI-3K/Akt and MEK/Erk pathways. Sci China Life Sci 58: 531-540 617 618 Cheyssac C, Dina C, Lepretre F, Vasseur-Delannoy V, Dechaume A, Lobbens S, Balkau B, Ruiz J, Charpentier G, 619 Pattou F, Joly E, Prentki M, Hansen T, Pedersen O, Vaxillaire M, Froguel P (2006) EIF4A2 is a positional candidate 620 gene at the 3q27 locus linked to type 2 diabetes in French families. Diabetes 55: 1171-1176 621 622 Chidlow G, Wood JP, Knoops B, Casson RJ (2016) Expression and distribution of peroxiredoxins in the retina and 623 optic nerve. Brain structure & function 221: 3903-3925 624 625 Cox J, Hein MY, Luber CA, Paron I, Nagaraj N, Mann M (2014) Accurate proteome-wide label-free quantification 626 by delayed normalization and maximal peptide ratio extraction, termed MaxLFQ. Mol Cell Proteomics 13: 2513- 627 2526 628 629 Cox J, Mann M (2008) MaxQuant enables high peptide identification rates, individualized p.p.b.-range mass 630 accuracies and proteome-wide protein quantification. Nature biotechnology 26: 1367-1372 631 632 Cox J, Neuhauser N, Michalski A, Scheltema RA, Olsen JV, Mann M (2011) Andromeda: a peptide search engine 633 integrated into the MaxQuant environment. Journal of proteome research 10: 1794-1805 634 635 De Ceuninck F, Dassencourt L, Anract P (2004) The inflammatory side of human chondrocytes unveiled by 636 antibody microarrays. Biochem Biophys Res Commun 323: 960-969 637 638 Hanschmann EM, Godoy JR, Berndt C, Hudemann C, Lillig CH (2013) Thioredoxins, glutaredoxins, and 639 peroxiredoxins--molecular mechanisms and health significance: from cofactors to antioxidants to redox signaling. 640 Antioxidants & redox signaling 19: 1539-1605 641 642 Harada A, Teng J, Takei Y, Oguchi K, Hirokawa N (2002) MAP2 is required for dendrite elongation, PKA 643 anchoring in dendrites, and proper PKA signal transduction. J Cell Biol 158: 541-549 644 645 He Y, Smith R (2009) Nuclear functions of heterogeneous nuclear ribonucleoproteins A/B. Cell Mol Life Sci 66: 646 1239-1256 647 648 Hollborn M, Tenckhoff S, Seifert M, Kohler S, Wiedemann P, Bringmann A, Kohen L (2006) Human retinal 649 epithelium produces and responds to placenta growth factor. Graefes Arch Clin Exp Ophthalmol 244: 732-741 650 22

651 Huang da W, Sherman BT, Lempicki RA (2009) Systematic and integrative analysis of large gene lists using 652 DAVID bioinformatics resources. Nature protocols 4: 44-57 653 654 Huang H, He J, Johnson D, Wei Y, Liu Y, Wang S, Lutty GA, Duh EJ, Semba RD (2015a) Deletion of placental 655 growth factor prevents diabetic retinopathy and is associated with Akt activation and HIF1alpha-VEGF pathway 656 inhibition. Diabetes 64: 200-212 657 658 Huang H, He J, Johnson D, Wei Y, Liu Y, Wang S, Lutty GA, Duh EJ, Semba RD (2015b) Deletion of placental 659 growth factor prevents diabetic retinopathy and is associated with Akt activation and HIF1alpha-VEGF pathway 660 inhibition. diabetes 2015;64:200-212. Diabetes 64: 1067 661 662 Iriuchijima N, Sato-Harada R, Takano M, Fujio K, Sato T, Goto F, Harada A (2005) Reduced expression of kinase- 663 associated phosphatase in cortical dendrites of MAP2-deficient mice. Biochem Biophys Res Commun 338: 1216- 664 1221 665 666 Ji J, Yang SN, Huang X, Li X, Sheu L, Diamant N, Berggren PO, Gaisano HY (2002) Modulation of L-type Ca(2+) 667 channels by distinct domains within SNAP-25. Diabetes 51: 1425-1436 668 669 Kanehisa M, Furumichi M, Tanabe M, Sato Y, Morishima K (2017) KEGG: new perspectives on genomes, 670 pathways, diseases and drugs. Nucleic Acids Res 45: D353-D361 671 672 Kowluru RA, Mishra M (2015) Contribution of epigenetics in diabetic retinopathy. Sci China Life Sci 58: 556-563 673 674 Lee R, Wong TY, Sabanayagam C (2015) Epidemiology of diabetic retinopathy, diabetic macular edema and related 675 vision loss. Eye and vision 2: 17 676 677 Li R, Shen Y (2013) An old method facing a new challenge: re-visiting housekeeping proteins as internal reference 678 control for neuroscience research. Life Sci 92: 747-751 679 680 Lim RW, Halpain S (2000) Regulated association of microtubule-associated protein 2 (MAP2) with Src and Grb2: 681 evidence for MAP2 as a scaffolding protein. J Biol Chem 275: 20578-20587 682 683 Liu J, Yeung PK, Cheng L, Lo AC, Chung SS, Chung SK (2015) Epac2-deficiency leads to more severe retinal 684 swelling, glial reactivity and oxidative stress in transient middle cerebral artery occlusion induced ischemic 685 retinopathy. Sci China Life Sci 58: 521-530 686 687 Luber CA, Cox J, Lauterbach H, Fancke B, Selbach M, Tschopp J, Akira S, Wiegand M, Hochrein H, O'Keeffe M, 688 Mann M (2010) Quantitative proteomics reveals subset-specific viral recognition in dendritic cells. Immunity 32: 689 279-289 690 691 Luttun A, Tjwa M, Moons L, Wu Y, Angelillo-Scherrer A, Liao F, Nagy JA, Hooper A, Priller J, De Klerck B, 692 Compernolle V, Daci E, Bohlen P, Dewerchin M, Herbert JM, Fava R, Matthys P, Carmeliet G, Collen D, Dvorak 693 HF et al (2002) Revascularization of ischemic tissues by PlGF treatment, and inhibition of tumor angiogenesis, 694 arthritis and atherosclerosis by anti-Flt1. Nat Med 8: 831-840 695 696 Ma Z, Portwood N, Foss A, Grill V, Bjorklund A (2005) Evidence that insulin secretion influences SNAP-25 697 through proteasomal activation. Biochem Biophys Res Commun 329: 1118-1126 698 699 Mohammed KA, Nasreen N, Tepper RS, Antony VB (2007) Cyclic stretch induces PlGF expression in bronchial 700 airway epithelial cells via nitric oxide release. American journal of physiology Lung cellular and molecular 701 physiology 292: L559-566 702 703 Oddoux S, Zaal KJ, Tate V, Kenea A, Nandkeolyar SA, Reid E, Liu W, Ralston E (2013) Microtubules that form 704 the stationary lattice of muscle fibers are dynamic and nucleated at Golgi elements. J Cell Biol 203: 205-213 705 706 Perumal N, Funke S, Pfeiffer N, Grus FH (2014) Characterization of lacrimal proline-rich protein 4 (PRR4) in 707 human tear proteome. Proteomics 14: 1698-1709 708 709 Perumal N, Funke S, Wolters D, Pfeiffer N, Grus FH (2015) Characterization of human reflex tear proteome reveals 710 high expression of lacrimal proline-rich protein 4 (PRR4). Proteomics 15: 3370-3381 711

712 Poch E, Giner V, Gonzalez-Nunez D, Coll E, Oriola J, de la Sierra A (2002) Association of the G protein beta3 713 subunit T allele with insulin resistance in essential hypertension. Clin Exp Hypertens 24: 345-353 714 715 Ponticelli S, Marasco D, Tarallo V, Albuquerque RJ, Mitola S, Takeda A, Stassen JM, Presta M, Ambati J, Ruvo M, 716 De Falco S (2008) Modulation of angiogenesis by a tetrameric tripeptide that antagonizes vascular endothelial 717 growth factor receptor 1. J Biol Chem 283: 34250-34259 718 719 Rocha-Martins M, Njaine B, Silveira MS (2012) Avoiding pitfalls of internal controls: validation of reference genes 720 for analysis by qRT-PCR and Western blot throughout rat retinal development. PLoS One 7: e43028 721 722 Schmidt O, Pfanner N, Meisinger C (2010) Mitochondrial protein import: from proteomics to functional 723 mechanisms. Nat Rev Mol Cell Biol 11: 655-667 724 725 Simo R, Hernandez C, European Consortium for the Early Treatment of Diabetic R (2014) Neurodegeneration in the 726 diabetic eye: new insights and therapeutic perspectives. Trends Endocrinol Metab 25: 23-33 727 728 Sodhi K, Maxwell K, Yan Y, Liu J, Chaudhry MA, Getty M, Xie Z, Abraham NG, Shapiro JI (2015) pNaKtide 729 inhibits Na/K-ATPase reactive oxygen species amplification and attenuates adipogenesis. Sci Adv 1: e1500781 730 731 Szklarczyk D, Franceschini A, Kuhn M, Simonovic M, Roth A, Minguez P, Doerks T, Stark M, Muller J, Bork P, 732 Jensen LJ, von Mering C (2011) The STRING database in 2011: functional interaction networks of proteins, 733 globally integrated and scored. Nucleic Acids Res 39: D561-568 734 735 Taipale M, Jarosz DF, Lindquist S (2010) HSP90 at the hub of protein homeostasis: emerging mechanistic insights. 736 Nat Rev Mol Cell Biol 11: 515-528 737 738 Teng J, Takei Y, Harada A, Nakata T, Chen J, Hirokawa N (2001) Synergistic effects of MAP2 and MAP1B 739 knockout in neuronal migration, dendritic outgrowth, and microtubule organization. J Cell Biol 155: 65-76 740 741 Tyanova S, Temu T, Sinitcyn P, Carlson A, Hein MY, Geiger T, Mann M, Cox J (2016) The Perseus computational 742 platform for comprehensive analysis of (prote)omics data. Nat Methods 13: 731-740 743 744 Van de Veire S, Stalmans I, Heindryckx F, Oura H, Tijeras-Raballand A, Schmidt T, Loges S, Albrecht I, Jonckx B, 745 Vinckier S, Van Steenkiste C, Tugues S, Rolny C, De Mol M, Dettori D, Hainaud P, Coenegrachts L, Contreres JO, 746 Van Bergen T, Cuervo H et al (2010) Further pharmacological and genetic evidence for the efficacy of PlGF 747 inhibition in cancer and eye disease. Cell 141: 178-190 748 749 Wild S, Roglic G, Green A, Sicree R, King H (2004) Global prevalence of diabetes: estimates for the year 2000 and 750 projections for 2030. Diabetes care 27: 1047-1053 751 752 Yang J, Gao J, Adamian M, Wen XH, Pawlyk B, Zhang L, Sanderson MJ, Zuo J, Makino CL, Li T (2005) The 753 ciliary rootlet maintains long-term stability of sensory cilia. Mol Cell Biol 25: 4129-4137 754 755 Yao YG, Yang HS, Cao Z, Danielsson J, Duh EJ (2005) Upregulation of placental growth factor by vascular 756 endothelial growth factor via a post-transcriptional mechanism. FEBS Lett 579: 1227-1234 757 758 Zhu L, Shen W, Zhu M, Coorey NJ, Nguyen AP, Barthelmes D, Gillies MC (2013) Anti-retinal antibodies in 759 patients with macular telangiectasia type 2. Invest Ophthalmol Vis Sci 54: 5675-5683 760 761

762 Figure legends

763 Figure 1: Graphical abstract of the animal breeding program (A); screening of diabetic conditions (B); retinal

764 sample collection (C), initial preparaition of peptide digests (D)

765

766 Figure 2: The overall workflow of our proteomics data analysis.

767 The mass spectral (MS) raw data were analyzed with MaxQuant computational proteomics platform (version

768 1.6.1.0) and its built-in Andromeda search engine. MS data was aligned against Mus musculus UniProtKB (83,598

769 entries) target database with collection of common contaminants and concatenated with the reversed versions of all

770 sequences. The FDR (false discovery rate) was set to 0.01. Label minimum ratio count was set to 2, peptides for

771 quantification was set to unique and razor and re-quantify to allow identification and quantification of proteins in

772 groups for LFQ analysis. Then for peptide identification, mass and intensity of the peptide peaks in mass

773 spectrometry (MS) spectrum are detected and assembled into three-dimensional (3D) peak hills over the m/z

774 retention time plane. Perseus analysis pipeline, functional classification and pathway analysis.

775

776 Figure 3: Functional annotation, reactome pathways and protein-protein interaction network of Akita.PlGF-/- vs

777 Akita. (A) Significantly found up and down-regulated proteins were represented as a heatmap. Green and red colour

778 indicates low to high intensity of abundance of proteins. (B) The differentially expressed proteins are involved in

779 different biological processes like biological regulation (GO:0065007), cellular component organization or

780 biogenesis (GO:0071840), cellular process (GO:0009987), localization (GO:0051179), metabolic process

781 (GO:0008152), multicellular organismal process (GO:0032501) and response to stimulus (GO:0050896)

782 respectively. (C) The differentially expressed proteins are involved in different molecular functions like antioxidant

783 activity (GO:0016209), binding (GO:0005488), catalytic activity (GO:0003824), signal transducer activity

784 (GO:0004871), structural molecule activity (GO:0005198) and transporter activity (GO:0005215) respectively. (D)

785 The differentially expressed proteins are involved in various cellular components functions like cell part

786 (GO:0044464), macromolecular complex (GO:0032991), membrane (GO:0016020), and organelle (GO:0043226)

787 respectively. (E) The differentially expressed proteins are different classes of proteins like chaperone, cytoskeletal

788 protein, enzyme modulator, hydrolase, membrane traffic protein, nucleic acid binding, oxidoreductase, transferase,

789 transporter and viral proteins respectively. (F) The differentially expressed proteins are involved in various

790 reactome biological pathways like G-protein activation pathway, Co-operation of PDCL (PhLP1) and TRiC/CCT in

791 G-protein beta folding pathway, Chaperonin-mediated protein folding pathway, Protein folding pathway, Regulation

792 of insulin secretion pathway, mRNA Splicing - Major Pathway and Transmission across Chemical Synapses

793 pathway respectively. (G) Highlighted the various colours of up and down-regulated proteins involved in complex

794 protein assembly (yellow) in biological process (GO:0006461), binding (blue) function of molecular function

795 (GO:0005488), membrane-bounded organelle (red) proteins in cellular component (GO:0043227) respectively.

796

797 Figure 4: Functional annotation, reactome pathways and protein-protein interaction network of Akita vs C57. (A)

798 Significantly found up and down-regulated proteins were represented as a heatmap. Green and red colour indicates

799 low to high intensity of abundance of proteins (B) The differentially expressed proteins are involved in different

800 biological processes such as biological regulation (GO:0065007), cellular component organization or biogenesis

801 (GO:0071840), cellular process (GO:0009987), developmental process (GO:0032502), localization (GO:0051179),

802 metabolic process (GO:0008152), multicellular organismal process (GO:0032501), response to stimulus

803 (GO:0050896) respectively. (C) The differentially expressed proteins are involved in various molecular functions

804 such as antioxidant activity (GO:0016209), binding (GO:0005488), catalytic activity (GO:0003824), structural

805 molecule activity (GO:0005198) and transporter activity (GO:0005215) respectively. (D) The differentially

806 expressed proteins are involved in various cellular components functions like cell part (GO:0044464),

807 macromolecular complex (GO:0032991), membrane (GO:0016020) and organelle (GO: 0043226) respectively. (E)

808 The differentially expressed proteins are different classes of proteins like cytoskeletal protein, enzyme modulator,

809 hydrolase, ligase, membrane traffic protein, nucleic acid binding, oxidoreductase and transporter respectively. (F)

810 The differentially expressed proteins are involved in various reactome biological pathways like Ion transport by P-

811 type, Ion channel transport, Transmembrane transport of small molecules, Ion homeostasis (Atp1a1, Atp1b2 and

812 Atp1a2), Cardiac conduction, Regulation of insulin secretion, Integration of energy metabolism, mRNA Splicing -

813 Major Pathway, Processing of Capped Intron-Containing Pre-mRNA, Transmission across Chemical Synapses,

814 Neuronal System and Unclassified pathways respectively. (G) Highlighted the various colours of up and down-

815 regulated proteins involved in nervous system development (GO:0007399) (red), response to glucose (GO:0009749)

816 (blue) of biological process, catalytic activity (GO:0003824) (light green), hydrolase activity (GO:0016787) (pink),

817 of molecular functions, Insulin secretion (ko04911) (yellow), Pancreatic secretion (ko04972) (cyan), Oxidative

818 phosphorylation (ko00190) (thick green) of KEGG pathways.

819

820 Figure 5: Functional annotation, reactome pathways and protein-protein interaction network of PlGF-/- vs C57. (A)

821 Significantly found up and down-regulated proteins were represented as a heatmap. Green and red colour indicates

822 low to high intensity of abundance of proteins (B) The differentially expressed proteins are involved in different

823 biological processes such as biological regulation (GO:0065007), cellular component organization or biogenesis

824 (GO:0071840), cellular process (GO:0009987), developmental process (GO:0032502), localization (GO:0051179),

825 locomotion (GO:0040011), metabolic process (GO:0008152), multicellular organismal process (GO:0032501),

826 response to stimulus (GO:0050896) respectively. (C) The differentially expressed proteins are involved in various

827 molecular functions such as binding (GO:0005488), catalytic activity (GO:0003824), signal transducer activity

828 (GO:0004871), structural molecule activity (GO:0005198), transporter activity (GO:0005215) respectively. (D) The

829 differentially expressed proteins are involved in various cellular components functions like cell part (GO:0044464),

830 macromolecular complex (GO:0032991), membrane (GO:0016020), organelle (GO:0043226), and synapse

831 (GO:0045202) respectively. (E) The differentially expressed proteins are classified in to various protein class such

832 as chaperone, cytoskeletal protein, enzyme modulator, hydrolase, membrane traffic protein, nucleic acid binding,

833 oxidoreductase, transcription factor, transferase, transporter, and viral protein respectively. (F) The differentially

834 expressed proteins are involved in various reactome biological pathways like Activation of the phototransduction

835 cascade, Signal Transduction, G-protein activation, Opioid Signalling, Signalling by GPCR, ADP signalling through

836 P2Y purinoceptor-12, Platelet activation, signalling and aggregation, Hemostasis, Cooperation of PDCL (PhLP1)

837 and TRiC/CCT in G-protein beta folding, Chaperonin-mediated protein folding, Metabolism of proteins, Activation

838 of BAD and translocation to mitochondria, Intrinsic Pathway for Apoptosis, Adrenaline, noradrenaline inhibits

839 insulin secretion, Regulation of insulin secretion, Metabolism, RHO GTPase Effectors, Signaling by Rho GTPases,

840 Glucagon-type ligand receptors, Activation of G protein gated Potassium channels, Transmission across Chemical

841 Synapses, Glucagon-like Peptide-1 (GLP1) regulates insulin secretion, Inhibition of voltage gated Ca2+ channels

842 via Gbeta/gamma subunits, Neuronal System, Inactivation, recovery and regulation of the phototransduction cascade,

843 Beta-catenin independent WNT signalling, Rho GTPase cycle and un classified pathway respectively. (G)

844 Highlighted the various colours of up and down-regulated proteins involved in nervous system development

845 (GO:0007399) (red), eye photoreceptor cell differentiation (GO:0001754) (blue) of biological process, hydrolase

846 activity (GO:0016787) (light green), catalytic activity (GO:0003824) (yellow) of molecular functions, photoreceptor

847 inner segment (GO:0001917) (pink), photoreceptor outer segment (GO:0001750) (thick green) of cellular

848 components, Ras signalling pathway (ko04014) (cyan), VEGF signalling pathway (ko04370) (orange), Oxidative

849 phosphorylation (ko00190) (magenta) of KEGG pathways.

850

851 Figure 6: Functional annotation, reactome pathways and protein-protein interaction network of Akita.PlGF-/- vs

852 PlGF-/-. (A) Significantly up and down-regulated proteins were represented as a heatmap. Green and red colour

853 indicates low to high intensity of abundance of proteins. (B) The differentially expressed proteins are involved in

854 different biological processes such as biological regulation (GO:0065007), cellular component organization or

855 biogenesis (GO:0071840), cellular process (GO:0009987), developmental process (GO:0032502), localization

856 (GO:0051179), metabolic process (GO:0008152), multicellular organismal process (GO:0032501), and response to

857 stimulus (GO:0050896) respectively. (C) The differentially expressed proteins are involved in various molecular

858 functions such as binding (GO:0005488), catalytic activity (GO:0003824), structural molecule activity

859 (GO:0005198), and transporter activity (GO:0005215) respectively. (D) The differentially expressed proteins are

860 involved in various cellular components functions like cell part (GO:0044464), macromolecular complex

861 (GO:0032991), membrane (GO:0016020) and organelle (GO:0043226) respectively (E) The differentially expressed

862 proteins are classified in to various protein class such as cell adhesion molecule, chaperone, cytoskeletal protein,

863 hydrolase, nucleic acid binding, oxidoreductase, receptor, signalling molecule, transcription factor, transferase and

864 transporter respectively. (F) The differentially expressed proteins are involved in various reactome biological

865 pathways like Attenuation phase, HSF1-dependent transactivation, Intraflagellar transport, Assembly of the primary

866 cilium, Organelle biogenesis and maintenance, Translocation of GLUT4 to the plasma membrane, Hedgehog 'off'

867 state, and Hedgehog 'on' state respectively. (G). Highlighted the various colours of up and down-regulated proteins

868 involved in protein folding (GO:0006457) (blue) of biological process, binding (GO:0005488) (red) of molecular

869 function, intracellular organelle part (GO:0044446) (green) of cellular component, Gap junction (ko04540) (green)

870 of KEGG pathway.

871

872 Figure 7: Comparison of each group of up and down-regulated proteins are illustrated by Venn diagrams. (A) The

873 three groups of unique up and down-regulated proteins are represented as a Venn diagram. (B) The four groups of

874 unique up and down-regulated proteins are represented as a Venn diagram.

875

876 877

878 Table Legends

879 Table 1: Proteomics samples of blood glucose, levels of glycated hemoglobin (HbA1c) and body weight.

880 Table 2: The list of significantly differentially expressed proteins, gene name, MS/MS count, unique sequence

881 coverage, molecular weight (MW), T-test difference, p-value, various KEGG name, and pathways in Akita.PLGF-/-

882 group compared to Akita group.

883

884 Table 3: The list of significantly differentially expressed proteins, gene name, MS/MS count, unique sequence

885 coverage, molecular weight (MW), T-test difference, p-value, various KEGG name, and pathways in Akita group

886 compared to C57 group.

887

888 Table 4: The list of significantly differentially expressed proteins, gene name, MS/MS count, unique sequence

889 coverage, molecular weight (MW), T-test difference, p-value, various KEGG name, and pathways in PLGF-/- group

890 compared to C57 group.

891

892 Table 5: The list of significantly differentially expressed proteins, gene name, MS/MS count, unique sequence

893 coverage, molecular weight (MW), T-test difference, p-value, various KEGG name, and pathways in Akita.PlGF-/-

894 group compared to PlGF-/- group.

895

896

897 Supplementary Figure Legend

898 Supplementary Figure 1: Quantitative profiling of retinal tissue between Akita.PlGF-/- vs. Akita groups. A.

899 Hierarchical clustering of Z-scored median LFQ intensities for all proteins in Akita.PlGF-/- vs. Akita groups. B.

900 LFQ intensity of each protein across Akita.PlGF-/- vs. Akita group samples. C. LFQ intensity of each protein across

901 Akita.PlGF-/- vs. Akita group samples after imputation. D. Pearson correlation coefficient of LFQ intensities among

902 LC-MS/MS runs in Akita.PlGF-/- vs. Akita groups.

903 Supplementary Figure 2: Quantitative profiling of retinal tissue between Akita vs. C57 groups. A. Hierarchical

904 clustering of Z-scored median LFQ intensities for all proteins in Akita vs. C57 groups. B. LFQ intensity of each

905 protein across Akita vs. C57 group samples. C. LFQ intensity of each protein across Akita vs. C57 group samples

906 after imputation. D. Pearson correlation coefficient of LFQ intensities among LC-MS/MS runs in Akita vs. C57

907 groups.

908 Supplementary Figure 3: Quantitative profiling of retinal tissue between PlGF-/- vs. C57 groups. A. Hierarchical

909 clustering of Z-scored median LFQ intensities for all proteins in PlGF-/- vs. C57 groups. B. LFQ intensity of each

910 protein across PlGF-/- vs. C57 group samples. C. LFQ intensity of each protein across PlGF-/- vs. C57 group samples

911 after imputation. D. Pearson correlation coefficient of LFQ intensities among LC-MS/MS runs in PlGF-/- vs. C57

912 groups.

913 Supplementary Figure 4: Quantitative profiling of retinal tissue between Akita.PlGF-/- vs PlGF-/- groups. A.

914 Hierarchical clustering of Z-scored median LFQ intensities for all proteins in Akita.PlGF-/- vs PlGF-/- groups. B.

915 LFQ intensity of each protein across Akita.PlGF-/- vs PlGF-/- group samples. C. LFQ intensity of each protein

916 across Akita.PlGF-/- vs. PlGF-/- group samples after imputation. D. Pearson correlation coefficient of LFQ

917 intensities among LC-MS/MS runs in Akita.PlGF-/- vs PlGF-/- groups.

918 Supplementary Figure 5: Representative mass spectra of A. Gnb1, B. Gnb2, C. Prdx6 and D. Map2 proteins.

30 bioRxiv preprint doi: https://doi.org/10.1101/338368; this version posted June 4, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Table 1: Proteomics samples of blood glucose, levels of glycated hemoglobin (HbA1c) and body weight.

Blood glucose Body weight Mouse strain HbA1c (%) (mg/dL) (g)

C57 141 6.5 27

163 7.3 26

170 7.6 22

157 7.1 23

Akita 628 23.5 22 616 23 23

439 16.9 26

670 25 23

PlGF-/- 138 6.4 32 161 7.2 30

123 5.9 31

151 6.9 32

Akita.PlGF-/- 504 19.2 24 406 15.8 22

539 20.4 19

664 24.8 23

Table 2: The list of significantly differentially expressed proteins, gene name, MS/MS count, unique sequence coverage, molecular weight (MW), T-test difference, p-value, various KEGG name, and pathways in Akita.PLGF-/- group compared to Akita group.

Unique Mol. Student's T-test Student's T-test Test Gene sequence MS/MS Protein names weight Difference statistic KEGG KEGG name names coverage count P-value [kDa] -/-_Akc57 -/-_Akc57 [%] AkiPLGF AkiPLGF 2.6501 ko00360 Methane metabolism; Peroxiredoxin-6 Prdx6 22.3 24.826 25 1.82912302 5.092553817 ko00680 Phenylalanine metabolism; ko00940 Phenylpropanoid biosynthesis Microtubule- 1.4644 ko04012 ErbB signaling pathway associated protein; ko04722 Neurotrophin signaling pathway Map2 9.7 49.264 46 0.723612309 2.727109127 Microtubule- ko04728 Dopaminergic synapse associated protein 2 ko04310 Wnt signaling pathway 1.3651 ko04145 Gap junction Tubulin beta-6 chain Tubb6 4 50.09 174 0.552398205 2.556084196 ko04540 Pathogenic Escherichia coli infection ko05130 Phagosome Rootletin Crocc 18.3 226.94 143 0.509349823 4.191491283 2.2411

2.0981 ko04141 Antigen processing and presentation ko04612 NOD-like receptor signaling pathway ko04621 Heat shock protein Pathways in cancer Hsp90ab1 8.3 83.28 83 0.483288288 3.900802892 ko04626 Plant-pathogen interaction HSP 90-beta ko04914 Progesterone-mediated oocyte maturation ko05200 Prostate cancer ko05215 Protein processing in endoplasmic reticulum Creatine kinase B- 1.3746 Ckb 62.7 42.713 311 0.36639595 2.572327037 ko00330 Arginine and proline metabolism type Guanine nucleotide- 1.4652 binding protein Gnb2 20.9 37.331 148 -0.180434704 -2.728495349 ko04062 Chemokine signaling pathway G(I)/G(S)/G(T) subunit beta-2 Synaptosomal- 1.3818 associated protein Snap25 71.4 23.315 276 -0.303587437 -2.584567189 ko04130 SNARE interactions in vesicular transport 25 Poly(rC)-binding 2.6793 Pcbp1 12.1 37.497 68 -0.339476585 -5.161235546 ko03040 Spliceosome protein 1 1.4059 ko04260 Aldosterone-regulated sodium reabsorption ko04960 Bile secretion ko04961 Carbohydrate digestion and absorption ko04964 Cardiac muscle contraction Sodium/potassium- ko04970 Endocrine and other factor-regulated calcium reabsorption transporting ATPase Atp1a1 25.8 112.98 395 -0.356846809 -2.626044796 ko04971 Gastric acid secretion subunit alpha-1 ko04972 Mineral absorption ko04973 Pancreatic secretion ko04974 Protein digestion and absorption ko04976 Proximal tubule bicarbonate reclamation ko04978 Salivary secretion 1.5331 ko00250 Alanine, aspartate and glutamate metabolism Poly(rC)-binding Vascular smooth muscle contraction Pcbp3 20.8 24.117 82 -0.35725832 -2.846988801 ko04270 protein 3 Focal adhesion ko04510 Regulation of actin cytoskeleton bioRxiv preprint doi: https://doi.org/10.1101/338368; this version posted June 4, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

ko04810 Oxytocin signaling pathway ko04921 1.3155 ko04730 Guanine nucleotide- Chagas disease (American trypanosomiasis) ko04916 Long-term depression binding protein G(o) Gnao1 37.6 40.084 201 -0.388885021 -2.47156149 ko05142 Melanogenesis subunit alpha ko05145 Toxoplasmosis Plasma membrane 1.7291 ko04020 Calcium signaling pathway calcium-transporting Atp2b1 4.7 134.75 48 -0.422261715 -3.197607557 ko04970 Pancreatic secretion ATPase 1 ko04972 Salivary secretion 1.6893 ko00190 ATP synthase Alzheimer's disease ko05010 Huntington's disease subunit O, Atp5o 37.1 23.363 79 -0.423535824 -3.125266485 ko05012 Oxidative phosphorylation mitochondrial ko05016 Parkinson's disease 1.7212 ko04020 ko04062 Amoebiasis ko04360 Axon guidance Calcium signaling pathway ko04530 Chagas disease (American trypanosomiasis) ko04540 Chemokine signaling pathway ko04670 Guanine nucleotide- Gap junction ko04730 Gastric acid secretion binding protein G(i) Gnai2 22.5 40.489 103 -0.461163998 -3.18320836 ko04740 Leukocyte transendothelial migration subunit alpha-2 ko04914 Long-term depression ko04916 Melanogenesis Olfactory transduction ko04971 Progesterone-mediated oocyte maturation ko05142 Tight junction ko05145 Toxoplasmosis ko05146 1.3480 ko05014 Amyotrophic lateral sclerosis (ALS) ko05410 Arrhythmogenic right ventricular Vimentin Vim 73.6 53.687 584 -0.532655716 -2.526809502 cardiomyopathy (ARVC) ko05412 Dilated cardiomyopathy ko05414 Hypertrophic cardiomyopathy (HCM) Mitochondrial 1.9014 Slc25a22 26.6 24.64 27 -0.689178944 -3.51807772 glutamate carrier 1 Heterogeneous nuclear Hnrnpa1 30.3 38.833 65 -0.78055954 -3.272488143 1.7700 ko03040 Spliceosome ribonucleoprotein A1 1.6172 ko05322 Systemic lupus erythematosus ko05034 Alcoholism Histone H1.2 Hist1h1c 21.2 21.266 115 -1.250981331 -2.995659744 ko04115 p53 signaling pathway ko04914 Progesterone-mediated oocyte maturation T-complex protein 1 1.8910 Cct8 10.8 53.082 21 -1.38865757 -3.498318032 subunit theta 1.6272 ko05322 Systemic lupus erythematosus Histone H1.4 Hist1h1e 16 21.977 119 -1.393318176 -3.013641476 ko05034 Alcoholism 1.7237 ko05322 Systemic lupus erythematosus Histone H1.3 Hist1h1d 9.5 22.099 105 -1.504905224 -3.187679564 ko05034 Alcoholism Heterogeneous 2.6967 nuclear Hnrnpc 28.8 34.384 51 -1.816102505 -5.20234665 ko03040 Spliceosome ribonucleoproteins C1/C2 V-type proton 1.4782 ko00190; Collecting duct acid secretion Atp6v1e1 27.9 26.157 29 -1.93034029 -2.750957456 ATPase subunit E 1 ko04145; Epithelial cell signaling in Helicobacter pylori bioRxiv preprint doi: https://doi.org/10.1101/338368; this version posted June 4, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

ko04966; infection ko05110; Oxidative phosphorylation ko05120; Phagosome Rheumatoid arthritis ko05323 Vibrio cholerae infection Heterogeneous 2.0245 nuclear Hnrnpul2 16.1 84.939 40 -1.97420311 -3.755246915 ribonucleoprotein U-like protein 2 1.9243 ko05016 Huntington s disease ko04919 Thyroid hormone signaling pathway Methyl-CpG- ko05202 Transcriptional misregulation in cancer Mecp2 19 52.307 22 -2.162194252 -3.561647541 binding protein 2 ko04330 Notch signaling pathway ko04110 Cell cycle ko04152 AMPK signaling pathway 1.8152 ko05322 Systemic lupus erythematosus Histone H1.5 Hist1h1b 15.7 22.576 26 -2.227953911 -3.3561123 ko04115 p53 signaling pathway Heterogeneous 1.7834 nuclear Hnrnpab 29.6 33.816 30 -2.522982121 -3.297190543 ko03040 Spliceosome ribonucleoprotein A/B Sodium- and 1.7807 chloride-dependent Slc6a11 7.7 69.96 28 -2.539692879 -3.292265714 GABA transporter 3 Heterogeneous 1.7754 nuclear Hnrnpd 22.7 24.611 50 -2.972328186 -3.282508326 ko03040 Spliceosome ribonucleoprotein D0 2.2317 04115 p53 signaling pathway 04110 Cell cycle 05322 Systemic lupus erythematosus Histone H1.1 Hist1h1a 23 21.785 26 -3.279434204 -4.172174855 05214 Glioma 05200 Pathways in cancer 04068 FoxO signaling pathway

Table 3: The list of significantly differentially expressed proteins, gene name, MS/MS count, unique sequence coverage, molecular weight (MW), T-test difference, p-value, various KEGG name, and pathways in Akita group compared to C57 group.

Unique Mol. Student's T-test Student's T-test Gene MS/MS Protein names sequence weight Difference Test statistic P-value KEGG KEGG name name count coverage [%] [kDa] Akc57_ c57 Akc57_ c57 Heterogeneous nuclear Hnrnpd 22.7 24.611 50 2.259608746 2.549069207 1.7485 ko05014 Amyotrophic lateral sclerosis (ALS) ribonucleoprotein D0 Heterogeneous nuclear ribonucleoproteins Hnrnpc 28.8 34.384 51 1.582131863 2.60181487 1.6488 ko03040 Spliceosome C1/C2 Heterogeneous nuclear Hnrnpab 29.6 33.816 30 1.428434372 2.629619451 1.4263 ribonucleoprotein A/B ko04260 Aldosterone-regulated sodium ko04960 reabsorption Bile secretion ko04961 Carbohydrate digestion and absorption ko04964 Cardiac muscle contraction Sodium/potassium- ko04970 Endocrine and other factor-regulated transporting ATPase Atp1b2 35.2 33.344 95 0.736261368 2.537781636 1.6958 ko04971 calcium reabsorption subunit beta-2 ko04972 Gastric acid secretion ko04973 Mineral absorption ko04974 Pancreatic secretion Protein digestion and absorption ko04976 Proximal tubule bicarbonate reclamation ko04978 Salivary secretion Heterogeneous nuclear ribonucleoprotein A1 Heterogeneous nuclear Hnrnpa1 30.3 38.833 65 0.686339378 3.301199097 3.4874 ko03040 Spliceosome ribonucleoprotein A1, N- terminally processed Lamin-B2 Lmnb2 31.7 67.317 126 0.636933804 4.644004221 1.7731 ko00190 Alzheimer's disease ATP synthase subunit O, ko05010 Huntington's disease Atp5o 37.1 23.363 79 0.630249023 3.515582565 1.9293 mitochondrial ko05012 Oxidative phosphorylation ko05016 Parkinson's disease Guanine nucleotide- ko04062; binding protein Chemokine signaling pathway Gnb1 45 37.377 334 0.518396854 2.488870095 1.9654 ko04742; Photo transduction G(I)/G(S)/G(T) subunit ko04744 Taste transduction beta-1 Poly(rC)-binding protein Pcbp1 12.1 37.497 68 0.486163139 6.353710521 1.5319 ko03040 Spliceosome 1 Guanine nucleotide- binding protein Gnb2 20.9 37.331 148 0.440536022 5.2404351 2.1103 ko04062 Chemokine signaling pathway G(I)/G(S)/G(T) subunit beta-2 ko00190 Collecting duct acid secretion ko04145 Epithelial cell signaling in Helicobacter V-type proton ATPase pylori infection Atp6v1a 31 68.325 160 0.366929054 2.55926315 1.8044 ko04966 catalytic subunit A Oxidative phosphorylation ko05110 Phagosome ko05120 Rheumatoid arthritis bioRxiv preprint doi: https://doi.org/10.1101/338368; this version posted June 4, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

ko05323 Vibrio cholerae infection

ko04260 Aldosterone-regulated sodium ko04960 reabsorption Bile secretion ko04961 Carbohydrate digestion and absorption ko04964 Cardiac muscle contraction Sodium/potassium- ko04970 Endocrine and other factor-regulated transporting ATPase Atp1a2 1.4 103.58 296 0.365712643 2.748964859 1.3519 ko04971 calcium reabsorption subunit alpha-2 ko04972 Gastric acid secretion ko04973 Mineral absorption ko04974 Pancreatic secretion Protein digestion and absorption ko04976 Proximal tubule bicarbonate reclamation ko04978 Salivary secretion ko04260 Aldosterone-regulated sodium ko04960 reabsorption Bile secretion ko04961 Carbohydrate digestion and absorption ko04964 Cardiac muscle contraction Sodium/potassium- ko04970 Endocrine and other factor-regulated transporting ATPase Atp1a1 25.8 112.98 395 0.314900398 2.454267753 1.7570 ko04971 calcium reabsorption subunit alpha-1 ko04972 Gastric acid secretion ko04973 Mineral absorption ko04974 Pancreatic secretion Protein digestion and absorption ko04976 Proximal tubule bicarbonate reclamation ko04978 Salivary secretion Heterogeneous nuclear Hnrnpk 47.5 48.562 218 0.207906246 3.248606535 1.3053 ko03040 Spliceosome ribonucleoprotein K Syntaxin-binding protein Stxbp1 34 67.568 257 0.181161404 2.533583266 1.4770 1 ko00250 Alanine, aspartate and glutamate ko00330 metabolism Glutamine synthetase Glul 56 42.119 380 -0.203300953 -3.336068814 1.3670 Arginine and proline metabolism ko00910 Nitrogen metabolism ko02020 Two-component system ko04360 Cofilin-1 Axon guidance; Cfl1 57.8 18.559 71 -0.602310181 -3.925211198 2.7127 ko04666 Fc gamma R-mediated phagocytosis Cofilin-2 ko04810 Regulation of actin cytoskeleton ko00360 Methane metabolism Peroxiredoxin-6 Prdx6 22.3 24.826 25 -0.915928364 -2.844815673 3.1471 ko00680 Phenylalanine metabolism ko00940 Phenylpropanoid biosynthesis haemoglo Haemoglobin subunit bin ko05143 African trypanosomiasis 62 15.112 115 -0.935765743 -3.640587253 1.3257 alpha alpha2 ko05144 Malaria Hba Microtubule-associated protein Map2 9.7 49.264 46 -0.977025986 -3.571238769 1.9001 Microtubule-associated protein 2 Epb4.1l3 Band 4.1-like protein3 10.5 97.596 31 -1.136313438 -3.278261317 2.4527 ko04530 Tight junction Epb41l3 Neurotrimin Ntm 13.9 34.954 16 -1.203859806 -7.346191294 1.7855 bioRxiv preprint doi: https://doi.org/10.1101/338368; this version posted June 4, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

ko00190 Alzheimer's disease ko04260 Cytochrome b-c1 Cardiac muscle contraction Uqcrq 18.3 9.7681 18 -1.340533257 -3.13707011 1.3544 ko05010 Huntington's disease complex subunit 8 ko05012 Oxidative phosphorylation ko05016 Parkinson's disease Glial fibrillary acidic Gfap 24.4 49.899 36 -1.42773056 -2.661065428 1.4080 protein Nme1 Nucleoside diphosphate ko00230 Purine metabolism Gm2039 29.9 14.092 33 -2.119158745 -3.052309265 1.3918 kinase A and B ko00240 Pyrimidine metabolism 0Nme2 Neurofilament medium Nefm 1.7 95.94 32 -3.49536705 -3.232984617 1.3610 ko05014 Amyotrophic lateral sclerosis (ALS) polypeptide

Table 4: The list of significantly differentially expressed proteins, gene name, MS/MS count, unique sequence coverage, molecular weight (MW), T-test difference, p-value, various KEGG name, and pathways in PLGF-/- group compared to C57 group.

Unique Mol. Student's T-test Student's T-test sequence MS/MS Protein names Gene name weight Difference Test statistic P-value KEGG KEGG name coverage count [kDa] PLGF-/- _ c57 PLGF-/- _ c57 [%] ko00190 NADH dehydrogenase Alzheimer's disease ko05010 Huntington's disease [ubiquinone] 1 alpha Ndufa8 37.2 19.992 26 1.762011528 8.051339072 3.7069 ko05012 Oxidative phosphorylation subcomplex subunit 8 ko05016 Parkinson's disease Cold-inducible RNA- Cirbp 44.2 18.607 34 1.621991158 2.831672565 1.5244 binding protein Retinol-binding Rbp3 18.6 134.48 75 1.191641808 4.157159713 2.2245 protein 3 Beta-soluble NSF Napb 26.2 33.557 39 0.918866158 2.819272591 1.5173 attachment protein Rootletin Crocc 18.3 226.94 143 0.856175423 2.886512978 1.5556 Guanine nucleotide- ko04062 Chemokine signaling pathway binding protein G(T) Gngt1 83.8 8.5278 63 0.757047653 3.613410071 1.9513 ko04744 Photo transduction subunit gamma-T1 Guanine nucleotide- Gnat1 binding protein G(t) Gnat2 46 39.966 257 0.733029842 3.139031115 1.6969 ko04744 Photo transduction subunit alpha-1 Gnat3 ko04260 Aldosterone-regulated sodium ko04960 reabsorption Sodium/potassium- ko04961 Bile secretion Carbohydrate digestion and transporting ATPase Atp1b2 35.2 33.344 95 0.659002304 2.53779854 ko04964 1.3544 absorption subunit beta-2 ko04970 Cardiac muscle contraction ko04971 Endocrine and other factor- ko04972 regulated calcium reabsorption bioRxiv preprint doi: https://doi.org/10.1101/338368; this version posted June 4, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

ko04973 Gastric acid secretion ko04974 Mineral absorption ko04976 Pancreatic secretion Protein digestion and ko04978 absorption Proximal tubule bicarbonate reclamation Salivary secretion Guanine nucleotide- ko04062 binding protein Chemokine signaling pathway Gnb1 45 37.377 334 0.637954712 2.876404603 1.5499 ko04742 Photo transduction G(I)/G(S)/G(T) ko04744 Taste transduction subunit beta-1 ko00190 Alzheimer's disease ATP synthase subunit ko05010 Huntington's disease Atp5o 37.1 23.363 79 0.623045921 3.108559973 1.6801 O, mitochondrial ko05012 Oxidative phosphorylation ko05016 Parkinson's disease Lamin-B2 Lmnb2 31.7 67.317 126 0.57357502 3.640287115 1.9653 ko00190 ATP synthase F(0) Alzheimer's disease ko05010 Huntington's disease complex subunit B1, Atp5f1 27.7 28.948 72 0.409056664 3.447880438 1.8642 ko05012 Oxidative phosphorylation mitochondrial ko05016 Parkinson's disease Guanine nucleotide- binding protein Gnb4 2.9 37.379 36 0.344211578 5.244639277 2.7145 ko04062 Chemokine signaling pathway subunit beta-4 Guanine nucleotide- binding protein Gnb2 20.9 37.331 148 0.33198595 2.675625039 1.4347 ko04062 Chemokine signaling pathway G(I)/G(S)/G(T) Gnb4 subunit beta-2 Carbon fixation pathways in ko00020 Aconitate hydratase, prokaryotes Aco2 24.6 85.462 123 0.329710484 3.052902023 1.6492 ko00630 Citrate cycle (TCA cycle) mitochondrial ko00720 Glyoxylate and dicarboxylate metabolism Non-POU domain- containing octamer- Nono 37.8 54.54 133 -0.165612221 -2.52636357 1.3477 binding protein ko04145 Gap junction Pathogenic Escherichia coli Tubulin beta-3 chain Tubb3 13.3 50.418 256 -0.256998539 -2.50070599 1.3326 ko04540 infection ko05130 Phagosome Phagosome Vesicle-associated ko04130 Vamp1 Salivary secretion membrane protein 2; ko04145 SNARE interactions in Vamp2 50.3 17.862 128 -0.329986095 -2.971921974 1.6038 Vesicle-associated ko04962 vesicular transport Vamp3 membrane protein 3 ko04970 Vasopressin-regulated water reabsorption 14-3-3 protein Ywhab 13.8 18.348 85 -0.359491825 -2.755515905 1.4808 ko04110 Cell cycle bioRxiv preprint doi: https://doi.org/10.1101/338368; this version posted June 4, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

beta/alpha;14-3-3 ko04114 Neurotrophin signaling protein beta/alpha, N- ko04722 pathway terminally processed Oocyte meiosis Serine/arginine-rich Srsf1 26.1 28.329 37 -0.450505733 -2.726145759 1.4639 ko03040 Spliceosome splicing factor 1 Srsf9 Heterogeneous nuclear ribonucleoproteins Hnrnpc 28.8 34.384 51 -0.602185249 -3.918784687 2.1071 ko03040 Spliceosome C1/C2 haemoglobin Haemoglobin subunit ko05143 African trypanosomiasis alpha 2; 62 15.112 115 -0.609102249 -2.938422415 1.5850 alpha ko05144 Malaria Hba 14-3-3 protein ko04110 Cell cycle gamma;14-3-3 protein Neurotrophin signaling Ywhag 30.8 28.302 160 -0.657001495 -3.486308246 1.8846 ko04114 gamma, N-terminally pathway ko04722 processed Oocyte meiosis Band 4.1-like protein3 Band 4.1-like protein Epb4.1l3 10.5 97.596 31 -1.216077328 -2.96828511 1.6018 ko04530 Tight junction 3, N-terminally Epb41l3 processed ko04260 Cardiac muscle contraction Tropomyosin alpha-1 Tpm1 Dilated cardiomyopathy 11 28.343 35 -1.237607479 -2.489495851 1.3261 ko05410 chain Tpm2 Hypertrophic cardiomyopathy ko05414 (HCM) Eukaryotic initiation factor 4A-II; Eukaryotic initiation Eif4a1 factor 4A-II, N- 6.6 46.402 28 -1.270507813 -3.457459917 1.8693 ko03013 RNA transport Eif4a2 terminally processed; Eukaryotic initiation factor 4A-I Adherens junction ko04010 Amyotrophic lateral sclerosis ko04062 (ALS) ko04145 Axon guidance B cell receptor signaling ko04310 Ras-related C3 pathway ko04360 botulinum toxin Bacterial invasion of epithelial Rac1 ko04370 substrate 1; cells Rac2 28 23.432 41 -1.556331158 -2.702360444 1.4502 ko04380 Chemokine signaling pathway Ras-related C3 Rac3 ko04510 Colorectal cancer botulinum toxin ko04520 Epithelial cell signaling in substrate 3 ko04620 Helicobacter pylori infection Fc epsilon RI signaling ko04650 pathway ko04662 Fc gamma R-mediated ko04664 phagocytosis ko04666 Focal adhesion bioRxiv preprint doi: https://doi.org/10.1101/338368; this version posted June 4, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

ko04670 Leukocyte trans endothelial ko04722 migration ko04810 MAPK signaling pathway Natural killer cell mediated ko04972 cytotoxicity ko05014 Neurotrophin signaling ko05100 pathway ko05120 Osteoclast differentiation ko05131 Pancreatic cancer ko05200 Pancreatic secretion ko05210 Pathways in cancer Phagosome ko05211 Regulation of actin ko05212 cytoskeleton ko05416 Renal cell carcinoma Shigellosis Toll-like receptor signaling pathway VEGF signaling pathway Viral myocarditis Wnt signaling pathway Glial fibrillary acidic Gfap 24.4 49.899 36 -1.846882343 -5.608623476 2.8632 protein Heterogeneous nuclear Hnrnpa0 12.1 30.53 23 -1.986347198 -4.562048267 2.4153 ribonucleoprotein A0 Neurofilament Amyotrophic lateral sclerosis Nefm 1.7 95.94 32 -3.633944035 -3.699355388 1.9958 ko05014 medium polypeptide (ALS) Rho-related BTB domain-containing Rhobtb3 2.1 69.207 44 -5.331884861 -3.050748316 1.6480 protein 3

Table 5: The list of significantly differentially expressed proteins, gene name, MS/MS count, unique sequence coverage, molecular weight (MW), T-test difference, p-value, various KEGG name, and pathways in Akita.PlGF-/- group compared to PlGF-/- group.

Unique Mol. Student's T-test Student's T-test Test Gene MS/MS Protein names sequence weight Difference statistic P-value KEGG KEGG name count names coverage [%] [kDa] AkiPLGF-/-_ PLGF-/- AkiPLGF-/-_ PLGF-/-

ko04145; Gap junction; Pathogenic Escherichia Tubulin beta-6 chain Tubb6 4 50.09 174 0.511187553 2.707577783 1.4532 ko04540; coli infection; ko05130 Phagosome ko04145; Gap junction; Pathogenic Escherichia Tubulin beta-3 chain Tubb3 13.3 50.418 256 0.422751904 2.53021607 1.3500 ko04540; coli infection; ko05130 Phagosome bioRxiv preprint doi: https://doi.org/10.1101/338368; this version posted June 4, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

Antigen processing and presentation; ko04141; NOD-like receptor ko04612; signaling pathway; ko04621; Pathways in cancer; Heat shock protein Plant-pathogen Hsp90ab1 8.3 83.28 83 0.342055798 2.509301016 1.3377 ko04626; HSP 90-beta interaction; ko04914; Progesterone-mediated ko05200; oocyte maturation; ko05215 Prostate cancer; Protein processing in endoplasmic reticulum ko04145; Gap junction; Pathogenic Escherichia Tubulin beta-2B chain Tubb2b 3.6 49.953 499 0.264726639 2.471480938 1.3155 ko04540; coli infection; ko05130 Phagosome ko04145; Gap junction; Pathogenic Escherichia Tubulin beta-5 chain Tubb5 17.1 49.67 566 0.258309364 2.611766276 1.3976 ko04540; coli infection; ko05130 Phagosome Antigen processing and presentation; ko03040; Endocytosis; ko04010; MAPK signaling ko04141; Heat shock cognate 71 pathway; Hspa8 30.8 70.87 444 0.156526566 2.95283795 1.5931 ko04144; Measles; kDa protein ko04612; Protein processing in ko05145; endoplasmic ko05162 reticulum; Spliceosome; Toxoplasmosis ko00190; Alzheimer's disease; ATP synthase subunit ko05010; Huntington's disease; Atp5o 37.1 23.363 79 -0.416332722 -2.562415443 1.3688 Oxidative O, mitochondrial ko05012; phosphorylation; ko05016 Parkinson's disease Microtubule- associated protein 1B; Map1b 4.6 270.25 27 -0.428781033 -2.496298553 1.3301 MAP1B heavy chain; MAP1 light chain LC1 MICOS complex Immt 10.2 83.899 62 -0.451266289 -2.506606186 1.3361 subunit Mic60 Heterogeneous nuclear Hnrnpm 34 77.648 99 -0.533585072 -3.107757295 1.6796 ko03040 Spliceosome ribonucleoprotein M Voltage-dependent ko04020; Calcium signaling pathway; anion-selective Vdac2 19.1 30.446 59 -0.678982258 -2.85774511 1.5392 ko05012; Huntington's disease; channel protein 2 ko05016 Parkinson's disease Transcriptional activator protein Pur- Pura 19.6 34.883 17 -1.35616684 -3.491899783 1.8876 alpha T-complex protein 1 Cct8 10.8 53.082 21 -1.527114391 -4.160793495 2.2262 subunit theta Cold-inducible RNA- Cirbp 44.2 18.607 34 -1.614155769 -2.464409392 1.3113 binding protein Alzheimer's disease; Cytochrome c oxidase Cox5b 14.7 13.847 19 -1.654263973 -3.035962803 1.6397 ko00190; Cardiac muscle bioRxiv preprint doi: https://doi.org/10.1101/338368; this version posted June 4, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

subunit 5B, ko04260; contraction; mitochondrial ko05010; Huntington's disease; ko05012; Oxidative phosphorylation; ko05016 Parkinson's disease Methyl-CpG-binding Mecp2 19 52.307 22 -1.850307941 -3.542110193 1.9140 protein 2 Retinoschisin Rs1 30.6 20.987 22 -2.150679111 -3.433042051 1.8563 ko00190; Alzheimer's disease; NADH dehydrogenase ko05010; Huntington's disease; [ubiquinone] 1 alpha Ndufa8 37.2 19.992 26 -2.18885231 -4.40456783 2.3423 Oxidative ko05012; subcomplex subunit 8 phosphorylation; ko05016 Parkinson's disease Enhancer of Erh 46.5 8.2033 13 -2.454516888 -2.978946933 1.6078 rudimentary homolog

bioRxiv preprint doi: https://doi.org/10.1101/338368; this version posted June 4, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. bioRxiv preprint doi: https://doi.org/10.1101/338368; this version posted June 4, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. bioRxiv preprint doi: https://doi.org/10.1101/338368; this version posted June 4, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. bioRxiv preprint doi: https://doi.org/10.1101/338368; this version posted June 4, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. bioRxiv preprint doi: https://doi.org/10.1101/338368; this version posted June 4, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. bioRxiv preprint doi: https://doi.org/10.1101/338368; this version posted June 4, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. bioRxiv preprint doi: https://doi.org/10.1101/338368; this version posted June 4, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.